- Email: [email protected]
Natural history of prostatism: Factors associated with discordance between frequency and bother of urinary symptoms
NATURAL HISTORY OF PROSTATISM: FACTORS ASSOCIATED WITH DISCORDANCE BETWEEN FREQUENCY AND BOTHER OF URINARY SYMPTOMS* STEVEN J. JACOBSEN, M.D., PH.D. CYNTHIAJ. GIRMAN, M.S. HARRYA. GUESS, M.D., PH.D. LAUREL A. PANSER, M.A., M.S.
CHRISTOPHER G. CHUTE, M.D., DRPH. JOSEPH E. OESTERLING, M.D. MICHAEL M. LIEBER, M.D.
From the Section of Clinical Epidemiology and the Department of Urology, Mayo Clinic and Foundation, Rochester, Minnesota, and the Department of Epidemiology, Merck Research Laboratories, Blue Bell, Pennsylvania.
ABSTRACT-The objective of this study was to assess the association between frequency and bother of urinary symptoms in a population-based cohort of men and to identify psychosocial factors that are related to reporting heightened or subdued bother relative to symptom frequency. The survey was conducted among men aged forty to seventy-nine years in Olmsted County, Minnesota, the baseline component of a prospective cohort study. Men were queried about the frequency of urinary symptom occurrence and the perceived bother associated with the symptoms. A regression analysis of American Urologic Association (AUA) bother scores on AUA frequency scores demonstrated a tight correspondence (r2 = 0.71). Men with bother scores greater than predicted from their frequency scores were more likely to have sought health care for their urinary symptoms than men whose bother was close to predicted (14 versus 5 percent, respectively). These men with heightened bother were older, poorer, more anxious, and had lower general psychologic well-being scores than the men whose bother was similar to that expected from their reported frequency. Men whose bother was lower than would be expected were less likely to have sought health care for urinary symptoms in the past year (3%) but were of similar age and socioeconomic status as compared with men whose bother was close to expected. These men, however, were more depressed than men whose bother was commensurate with reported frequency. While the men who reported greater bother than expected from their symptom frequency were more likely to have sought medical care for urinary symptoms in the past year, it is not clear whether this greater healthcare-seeking behavior is because bother captures an additional component of urologic disease or is a manifestation of psychosocial differences.
Benign prostatic hyperplasia (BPH) is a common disease of men that is creating an increasing demand on the health care system.l In the United States, it has been estimated that 1.7 million physician office visits per year are related to BPH2 and, in 1989, approximately 340,000 men under*Tkis study was funded, in part, by research grantsfrom Merck Reseanh laboratories as part of the BPH Natural History Study Group and the National Institutes of Health (AR30582). Submitted: March 26, 1993, accepted (with revisions): June 1, 1993
UROLOGY /
DECEMBER 1993
/ VOLUME42, NUMBER6
went a transurethral prostatectomy3 Results from the Normative Aging Study have suggested that in the United States a forty-year-old man has a 30 percent chance of undergoing prostatectomy for BPH in the following forty years of life.4 The probability of treatment for BPH is likely to increase as more men may seek newly available medical therapy for their symptoms in order to avoid surgical management. While BPH is relatively common among older men, individual perceptions about the severity of
663
disease may vary. Previous research has noted that age-adjusted rates of prostatectomy for BPH vary considerably both between and within countries5-a stimulating the search for factors that might explain this tremendous small-area variation. Consequently, the focus of much of the health services research in urology has been on regional differences in practice patterns. Patterns in patient perceptions of urinary symptoms and the influence of these perceptions on the decision to seek care, however, have received much less attention. Nonetheless, treatment guidelines for BPH are currently being developed that take into account the frequency with which urinary symptoms occur. While it could be argued that the frequency of occurrence better defines the presence of disease, a recent study from this group suggested that the perceived bother from urinary symptoms is, in fact, a stronger predictor of health-care-seeking behavior for urinary dysfunction than is the frequency of symptom occurrence.9 This finding that bother from symptoms is more closely linked to health-care-seeking behavior than frequency seems straightforward, but little is known about why some men are bothered more (or less) by their urinary symptoms compared with the frequency with which symptoms occur. The Olmsted County Study of Urinary Symptoms and Health Status Among Men was designed, in part, to aid our understanding of the relationship between frequency and bother of urinary symptoms. During the baseline component, men from this population-based sample were queried about the frequency with which specific urinary symptoms occur as well as the perceived bother from these same symptoms. In this report, we describe the association between the frequency and bother of urinary symptoms. Further, we assess the relationship of several socioeconomic and psychometric characteristics with bother that is either subdued or heightened relative to reported frequency METHODS The Olmsted County Study of Urinary Symptoms and Health Status Among Men is a population-based, prospective cohort study initiated in December 1989 to study the natural history of benign prostatic hyperplasia. Many of the details of this study have been published previously.9v10 Briefly, this study was designed to measure the age-specific prevalence of urinary symptoms in a community-based sample of men and to follow these men over time for the onset as well as pro-
664
gression or regression of disease. Baseline symptom data were collected by questionnaire, and all study participants were asked to void into a standard urometer to determine the peak urinary flow rate. In addition, a probability sample of study participants was invited to a detailed urologic examination, which included a digital rectal examination and transrectal ultrasonography to estimate prostate size. The study protocol was reviewed and approved by the Mayo Institutional Review Board. Potential subjects included men aged forty to seventy-nine years who were residents of Olmsted County, Minnesota. The resources of the Rochester Epidemiology Project” were used to establish a sampling frame of Olmsted County residents. Men aged forty to seventy-nine years on January 1, 1990 were randomly selected within five-year age groups and two geographic strata (city of Roth es t er versus the balance of Olmsted County) at a 14 percent sampling fraction. After screening a potential subject’s medical history for exclusion and eligibility criteria,lO he was contacted and his participation in the study solicited. The exclusion criteria were established to remove men who had previously received surgical treatment for genitourinary conditions or had neurologic conditions thought to interfere with normal voiding. A small group of men with billing disputes or who were incarcerated in the Federal Medical Prison was also eliminated. Each participant was administered a structured, in-person interview eliciting information about current daily medications and family history of urologic disease by a trained field research assistant. After the interview, each man was asked to void into a standard urometer in privacy Finally, each man was administered a previously validated questionnaire12 while the research assistant remained in the home. This self-administered questionnaire included 134 elements about severity of and bother from urinary symptoms, health status, and several sociodemographic characteristics. There were twelve questions that elicited information about the frequency of specific urinary symptoms during the past month and an additional twelve questions queried the perceived bother of these same symptoms. Symptom frequency questions were measured on a 7-point scale from 0 (I do not have symptoms) to 6 (always); symptom bother questions were asked on a similar 7-point scale from 0 (no symptom) to 6 (extremely bothersome). Subjects were also queried about the degree of interference due to urinary problems for seven daily
UROLOGY
/ DECEMBER 1993 / VOLUME42, NLJMBER 6
activities including: drinking fluids before travel, drinking fluids before going to bed, driving for two hours without stopping, getting enough sleep at night, going places that may not have a toilet, playing sports outdoors, and going to movies, shows, church, etc. Responses to each of these activities were placed on a 5-point scale from 1 (none of the time) to 5 (all of the time). A composite score for symptom frequency was calculated to approximate the AUA symptom index for BPH13 by resealing our data in consultation with the AUA index developers as previously reported.9 Similarly, a composite score was calculated for bother to approximate the AUA bother score that is in development. In addition to these questions about urinary symptoms, the questionnaire included a subset of the General Psychologic Well-being Scale. 12s14 This scale consists of ten elements which are summed for an indication of general psychologic well-being. These elements also can be grouped into five subscales: anxiety, depression, positive well-being, self control, and vitality All ten elements are measured on a 6-point semantic differentiation scale from 1 to 6 with lower numbers representing more-impaired well-being. Finally, men were queried about several sociodemographic characteristics, including age, education, income, marital status, and employment status. METHODSOF ANALYSIS
For each of the twelve symptoms, a Spear-man rank-order correlation coefficient was calculated to measure the association between symptom bother and symptom frequency In order to classify a man according to the agreement between his frequency of urinary symptoms and their perceived bother, we compared a man’s AUA bother score with his AUA frequency score. Because these two scores are based on different scales (O-27 and O-34 for bother and frequency, respectively), we constructed a least squares regression model of AUA bother score on AUA frequency score. The coefficient of determination (r2> calculated from this model describes the amount of variance in bother score that is explained by variance in the frequency score. Further, the slope and intercept derived from the regression model predict an AUA bother score from the reported AUA frequency score, taking into account the scale differences and any overall downplaying of symptoms. The comparison of a man’s reported AUA bother score to the bother score predicted by the regression equation gives an indication of the
UROLOGY
/ DECEMBER1993
/ VOLUME42, NUMBER6
agreement between the frequency of urinary symptoms and their perceived bother. While the degree of departure from the line with which to label men as having heightened or subdued bother is somewhat arbitrary, for this study we consider a man whose observed bother score is more than 1 standard deviation (root mean square error) greater than his predicted bother score to have a heightened bother; whereas a man whose observed bother score was at least 1 standard deviation less than the predicted bother score is considered to have subdued bother. Men whose bother scores are within one standard deviation of their predicted bother score are considered to be concordant for symptom frequency and symptom bother. Similar cutpoints at 1.5 standard deviations led to similar results and only the former are presented. Based on the aforementioned classification, the distribution of demographic and psychosocial characteristics was compared among the three groups of men. A chi-square test of homogeneity was used to assess differences in proportions of factors among the three groups. Logistic regression models were constructed to simultaneously adjust for these demographic and psychosocial factors in predicting discordance between bother and frequency of symptoms. Models were constructed separately for men with heightened bother compared with men whose bother was close to predicted, and for men with subdued bother compared with men whose bother was close to predicted. Both backward- and forwardstepped algorithms were used to select the best set of predictor variables for each of the two models. Odds ratios were derived from the logistic regression models by exponentiating the estimated regression coefficients. For all statistical tests, a nominal p-value of co.05 was considered statistically significant. It should be recognized, however, that these p-values (and confidence intervals) do not take into account the multiple comparisons. All analyses were conducted using the SAS statistical packages (SAS Institute, Cary, NC). RESULTS Overall, the majority of men reported bother scores that were within one point of frequency scores for each of the twelve symptoms queried (Table I). Men were most concordant for dysuria, with 98 percent reporting a bother score within one point of frequency score, and least concordant for nocturia, with 85 percent of men being concordant. Of the men who were discordant, proportionately more reported bother scores at least 1 665
TABLE I.
Bother score relative to frequency for individual urinary symptoms (Olmsted County Study of Urinary Symptoms and Health Status Among Men) Bother Relative to Frequency’ Greater Bother n (4%)
Lower Bother n (%)
Similar n (%)
spearman Correlation
wet clothes
101 (4.8)
49 (2.3)
1949 (92.0)
0.80
Urgency
56 (2.6)
160 (7.6)
1900 (89.7)
0.73
Dribbling
42 (2.0)
202 (9.5)
1872 (88.3)
0.76
Repeat within 10 minutes
38 (1.8)
56 (2.6)
2022 (95.4)
0.05
Nocturia
36 (1.7)
252 (11.9)
1811 (85.5)
0.75
Incomplete empty
27 (1.3)
86 (4.1)
2003 (94.5)
0.71
Hesitancy
27 (1.3)
81 (3.8)
2007 (94.7)
0.71
Weak stream
26 (1.2)
254 (12.0)
1836 (86.6)
0.68
stop/start
25 (1.2)
151 (7.1)
1940 (91.6)
0.69
Daily frequency
24 (1.1)
265 (12.5)
1827 (86.2)
0.65
Dysuria
14 (0.7)
17 (0.8)
2085 (98.4)
0.66
Strain
15 (0.7)
71 (3.4)
2030 (95.8)
0.67
Symptomspresentedin otder ojdectraring magnitudeofproportion with heightened symptomawamness. Row totals may differdue to single item non-response. *GreaterBother-Bother scan at Ieast two greater thanjrequencyscan; Lower Bother-Bother score at least two less thanjrequencyscore;Similar-Bother scwe within one ojjrequencyscow
30:
AlJAmFIGURE 1. AUA bother score by AUA frequency score: Olmsted County Study of Urinary Symptoms and Health Status Among Men. Lines represent least-squares regression and -+1 standard deviation about the least-squares regression line. Bubble size is proportionate to number of men represented by the point.
666
UROLOGY
/ DECEMBER1993
I VOLUME42, NUMBER6
TABLE II.
Demographic and socioeconomic characteristics of men by concordance of observed and predicted urinary symptom bother [Olmsted County Study of Urinary Symptoms and Health Status Among Men) Heightened Bother’ (n=242) n (96)
Subdued Bother’ (n=266) n (4%)
Concordant (n= 1608) n (%)
~65
157 (65)
208 (78)
1279 (80)
265
85 (35)
58 (22)
329 (20)
44 (18)
25 (10)
154 (10)
High School Graduate
61 (26)
51 (20)
340 (21)
> High School Graduate
133 (56)
184 (71)
1099 (69)
<25
79 (34)
49 (20)
309 (20)
2545
73 (31)
75 (30)
475 (31)
245
83 (35)
126 (50)
751 (49)
Full-time
125 (48)
170 (64)
1089 (68)
115 (52)
95 (36)
512 (32)
Married
210 (87)
247 (93)
1436 (90)
Other
31 (13)
19 (7)
167 (10)
Characteristic Age, yrt
Education’
Income, $ (1,000’s)*
Current JQnploymentt
Marital Status
Health-CarAeeking
Behavioti
YeS
34 (14)
8 (3)
86 (5)
No
205 (86)
253 (97)
1505 (95)
Columntotals may diner due to single item non-response. *Heightenedbother;ObservedALJAbotherscore more than one standarddeviation above pwdicted bother scorefrom regressionof AUA bother score on AlJAfquency score. tSubdwd bother; ObservedAUA bother SCORmore than one standarddeviation below predictedbother score from regressionof AUA botherSCOH on AUAfrequcncyscore. +x2 test, p < 0.001. §x’ test, p = 0.05.
lower than frequency score for every symptom except for wet clothing. Overall, 259 men reported a greater bother than frequency score for at least one symptom, 930 reported a lower bother than frequency score for at least one symptom, and 1,030 men reported bother scores within 1 point for all twelve symptoms (116 men reported lower bother for at least one symptom and greater bother for another symptom). Each man’s AUA bother score is plotted against his AUA frequency score in Figure 1. AUA bother scores were very closely related to AUA frequency
UROLOGY
I DECEMBER1993
! VOLUME42, NUMBER6
scores, with a coefficient of determination of 0.71. Based on the regression analysis, 1,608 men (76%) had bother scores within one standard deviation of their bother score predicted on the basis of AUA frequency score (between upper and lower lines, Fig. 1) Two hundred sixty-six men (13%) reported subdued bother, with bother scores more than one standard deviation below their predicted bother score; 242 men (11%) reported heightened bother, with bother scores more than one standard deviation above their predicted bother score.
667
TABLE III.
Psychologic well-being of men by concordance of observed and predicted urinary symptom bother [Olmsted County Study of Urinary Symptoms and Health Status Among Men) Scale
Heightened Bother’ SubduedBother’ (n=242) (n=266) n (%) n (W
ConcordantBother (n= 1608) n (W
General Well-Bei& 240
208 (86)
237 (90)
1512 (94)
<40
34 (14)
27 (10)
93 (6)
r 10
174 (82)
202 (77)
1321 (82)
< 10
68 (18)
60 (23)
283 (18)
2 10
239 (99)
257 (98)
1586 (99)
< 10
3 (1)
5 (2)
14 (1)
r5
198 (82)
232 (88)
1432 (89)
<5
44 (14)
33 (12)
173 (11)
210
214 (88)
239 (91)
1514 (95)
< 10
28 (12)
24 (9)
87 (5)
210
194 (80)
222 (84)
1445 (90)
< 10
48 (20)
42 (16)
160 (10)
Anxiety
Positivity
Self-Control*
Vitality*
Depression*
For all scores, a higher score represents betterfunction. Column totals may di@ due to single item non-response. ‘Heightened bother: Observed AUA bother score more than one standard deviation above predicted bother scoreJrom regression of AUA bother score on AUA Jrequency score. tsubdued bother:Observed AUA bother scow more than one standanldeviation below predicted bother scoreJrom regression of AUA bother score on AUA frequency score. Q2 test of homogeneity, p < 0.01.
As would be expected, men with heightened bother were much more likely to have sought health care in the past year for their urinary symptoms than men with subdued bother or men whose observed and predicted bother scores were concordant (14% versus 3% and 5%, respectively). Men with heightened bother were older, less educated, less likely to be employed full-time, less likely to be married, and had lower incomes than men whose observed and predicted bother score were close to one-another (Table II). In contrast, men with subdued bother were similar to the men with concordant observed and predicted bother scores in nearly all these characteristics. Men with heightened bother also had lower psychologic well-being scores than men who were concordant
668
(Table III), overall and for each of the subscales. Men with subdued bother show similar trends, with lower psychologic well-being scores for overall general well-being, anxiety, depression, and vitality Scores for each of these scales, however, were intermediate to those for men with heightened bother and men who were concordant. After simultaneous adjustment for all factors using logistic regression, men with heightened bother remained more likely to be older, have lower income, have lower general psychologic well-being scores, and be more anxious than men who were concordant for urinary symptom bother and frequency (Table IV). For men with subdued bother, however, none of the demographic or socioeconomic factors remained significant in the
UROLOGY
/ DECEMBER1993
/ VOLUME42, NKMBER6
TABLE IV.
Multiple logistic regression of psychosocial characteristics with heightened or subdued bother (Olmsted County Study of Urinary Symptoms and Health Status Among Men) Heightened Bother’ Crude*
Adjusted’
Subdued Bother+ Crude
Adjusted
4% yr c65’
1.0
1.0
1.0
265
2.1 (1.6, 2.8)’
2.0 (1.4, 2.8)
1.1 (0.8, 1.5)
__
Education
2.4 (1.6, 3.4)
High school graduate
1.5 (1.1, 2.1)
__
1.0 (0.6, 1.5) 0.9 (0.6, 1.2)
1.0
__
1.0
<25
2.3 (1.7, 3.2)
1.7 (1.2, 2.5)
1.0 (0.7, 1.4)
25-45
1.4 (1.0, 1.9)
1.3 (0.9, 1.8)
1.0 (0.7, 1.3)
2451
1.0
1.0
1.0
1.0
__
> High school graduate1
__ __
Income, 1,OOOsof dollars/yr
__ __
Current Employment Status < Full-time1 Full-time
1.0 0.8 (0.6, 1.1)
0.5 (0.4, 0.7)
__ __
Marital Status 1.3 (0.8, 1.9)
__ __
0.7 (0.4, 1.1)
2.7 (1.8, 4.0)
1.9 (1.1, 3.3)
1.9 (1.2, 2.9)
1.0
1.0
1.0
No (2 10)’
1.0
1.0
1.0
Yes (C 10)
1.8 (1.3, 2.5)
1.7 (1.2, 2.5)
1.4 (1.0, 1.9)
Married1
1.0
Other
1.0
__ __
General Psychologic Well-Being C40 r40’
__ __
Anxiety
__ __
Positivity Yes (z-10)1 No (< 10)
1.0
1.0 1.4 (0.4, 5.0)
__
2.2 (0.8, 6.2)
1.0
__ __
1.2 (0.8, 1.8)
__ __
Self-Control Yes (z-5)’ No (
1.8 (1.3, 2.6)
1.0
__ __
Vitality Yes (ZlO)l No (< 10)
1.0
1.0
__ -_
2.3 (1.5, 3.6)
__
1.8 (1.1, 2.8)
1.0
__ __
1.0
1.0
1.7 (1.2, 2.5)
1.7 (1.1, 2.4)
Depression No (5: 10)’ Yes (
2.2 (1.6. 3.2)
*Heightenedbothn: Observed AUA botherscore more than one standarddeviationabove predictedbotherscorefrom regressionof AUA bother scan on AUAfreqwncy score. tSubduedbother:ObservedAUA botherscore more than one standurddeviationbelow predictedbotherscorefrom regressionof AUA bother score on AUAfrequencyscow. iBiva7iate association. §Adjusted_for other&ton in model (factorsp > 0.10 not includedin model). llRejerence category. POddsratio (95%confidenceinterval).
UROLOGY
/ DECEMBER1993
/ VOLUME42, NUMBER6
669
multivariable model; only depression was predictive of subdued bother relative to concordance. COMMENT In this community-based cohort of men, the frequency with which urinary symptoms were reported corresponded very closely to reported bother from urinary symptoms. There were, however, a number of men whose urinary symptoms were more bothersome than would be expected from their frequency of occurrence. These men with heightened bother tended to be older, poorer, and more anxious than men whose bother from urinary symptoms was comparable to the reported frequency. Similarly, there were a number of men who tended to downplay their urinary symptoms (with bother scores low relative to frequency scores). These men tended to be more depressed than men whose bother was comparable to their frequency of urinary symptoms, but were no different in terms of age, education, or income. The presence of heightened bother from urinary symptoms was reflected in health-care-seeking behavior; men with heightened bother were much more likely to have sought medical care in the past year for urinary symptoms than men whose bother was comparable or subdued relative to their urinary symptom frequency While the majority of men reported bother scores that were comparable to their frequency scores, there was, in general, a downward scoring of symptom bother relative to frequency. Obviously, a portion of this difference is attributable to differences in scales; the range of possible scores is smaller for bother than for frequency In addition, some portion of this downward scoring could be due to other factors. These could include a reluctance to admit to bother (a general stoicism) or, alternatively, some degree of acceptance. Unfortunately, we are unable to determine how much of the downward scoring is due to scale and how much is due to other factors. Regardless, as treatment recommendations are suggested to be based on frequency of urinary symptoms, it is reassuring that the amount of bother reported by most men in this community-based sample is in accord with their reported frequency of urinary symptoms. Among the men whose bother seems disproportionately greater than their reported frequency of urinary symptoms, there is a strong association with health-care-seeking behavior for urinary symptoms. The appropriateness of this heightened bother, however, is not easily discernible. The higher likelihood of seeking health care may be
due to clinical disease that is worse than indicated by frequency scores. That is, bother may add an additional dimension of disease that is not captured by frequency of symptom occurrence which should be considered in treatment decisions. Alternatively, heightened bother may not be related to worse clinical disease, but instead to other psychosocial factors.‘* The association with older age may be due to older men having more free time to contemplate symptoms than younger men. The younger men might also be more likely to experience social pressures to downplay bother.16 Further, older men may have been more likely to have grown weary of their symptoms and are at an age when acute changes might be more likely to be noticed and perceived as abnormal.r7 The association between heightened bother and anxiety is particularly interesting. If the heightened bother was due to anxiety, then perhaps men with heightened bother should be channeled into treatments that would address their anxiety and other psychologic aspects of their disease as opposed to surgical treatments. Unfortunately, these crosssectional data cannot distinguish the causal pathway In contrast to men with heightened bother, there was no association between age and subdued bother. This suggests that there is no additional acceptance of bother from symptoms’s with age beyond that accounted for by the regression of bother on frequency score. The association between depression and subdued bother suggests, however, an appealing causal hypothesis, that depressed men are reluctant to admit bother or have other, more pressing concerns. In the interpretation of these results, several potential limitations must be kept in mind. As mentioned, the cross-sectional nature of these data precludes the understanding of the temporal relationships between onset of symptoms, progression or regression of symptoms, and the association with onset of bother. The longitudinal follow-up of this cohort, however, should provide us with this information in the future. The exclusion of men who had already been surgically treated for their urinary symptoms may have selected out the most severely diseased men from the cohort. If these excluded men were more likely to have heightened bother, the association between heightened bother and health-care-seeking behavior in this study was most likely an underestimate of the true association. The 55 percent response rate must also be considered. If men who refused to participate were systematically more likely to have sought health care in the past year and were
UROLOGY
/ DECEMBER 1993 / VOLUME42, NUMBER6
more likely to report similar frequency and bother scores, our results might be biased. Further, respondents may differ from non-respondents systematically in the distribution of psychosocial parameters, providing the chance for further bias. These complex interactions, however, would probably not be strong enough to explain our findings. Finally, it must be recognized that multiple statistical tests and estimates have been calculated. Consequently, true p-values are greater than the nominal values presented and 95 percent confidence are wider than shown. The point estimates of the associations, however, remain unchanged. In conclusion, we found that while most men report a degree of bother associated with urinary symptoms that is in accord with the frequency of the symptoms, there are men in the community whose symptoms bother them more than the frequency with which their symptoms occur would suggest. These men with heightened bother appear to be more prone to have sought medical care for urinary symptoms in the past year, but it is not clear that this health-care-seeking behavior was inappropriate. As we follow these men over time, we should be able to sort out the appropriateness of this increased health-care-seeking behavior and develop a better understanding of BPH as a disease entity. Steven J. Jacobsen, M.D. DepartmentofHealth Sciences Research Mayo Clinic 200 First Street, SW Rochestel; Minnesota 55905 ACKNOWLEDGMENT. We thank the Ohnsted County Study of Urinary Symptoms and Health Status Among Men research staff, including Rebecca Nelson, Roger Mueller, and Charles Wilson for their assistance in the conduct of this study. In addition, we sincerely thank Ms. Sondra Buehler for her adept preparation of this manuscript. Finally, we wish to express our sincere appreciation of the enthusiasm of all members of the Ohnsted County Study of Urinary Symptoms and Health Status Among Men, without whose cooperation this study wouldnot be possible. , REFERENCES 1. Guess HA: Benign prostatic hyperplasia: antecedents and natural history. Epidemiol Rev 14: 131-153, 1992. 2. National Center for Health Statistics: United States Life Tables. U.S. Decennial Life Tables for 1979-81. Vol. 1, No.1. DHHS Pub. No. (PHS) 85-1150-1, Washington, D.C., U.S. Government Printing Office, 1985. 3. Graves EJ: Detailed Diagnoses and Procedures, National Hospital Discharge Survey, 1989. Vital Health Statistics 13( lOf3), Hyattsville, MD, National Center for Health Statistics, 199 1.
UROLOGY
/ DECEMBER 1993 / VOLUME42, NLIMBER6
4. Glynn RJ, Campion Ew, Bouchard GR, and Silbert JE: The development of benign prostatic hyperplasia among volunteers in the Normative Aging Study. Am J Epidemiol 121: 78-90,1985. 5. Wennberg JE, Mulley AG Jr, Hanley D, Timothy RP, Fowler FJ Jr, Roos NP. Barry MJ, McPherson K, Greenberg ER, Soule D, et al.: An assessment of prostatectomy for benign urinary tract obstruction. Geographic variations and the evaluation of medical care outcomes. JAMA 259: 3027-3030, 1988. 6. Wennberg J, and Gittelsohn A: Small area variations in health care delivery, Science 182: 1102-l 108, 1973. 7. McPherson K, Wennberg JE, Hovind OB, and Clifford P: Small-area variations in the use of common surgical procedures: an international comparison of New England, England, and Norway. N Engl J Med 307: 1310-1314,1982. 8. Stephenson Wp, Chute CG, Guess HA, Schwartt S, and Lieber M: Incidence and outcome of surgery for benign prostatic hyperplasia among residents of Rochester, Minnesota: 1980-87. A population-based study. Urology (Suppl 1) 38: 3242,199l. 9. Jacobsen SJ, Guess HA, Panser L, Girman CJ, Chute CG, Oesterling JE, and Lieber MM: A population-based study of healthcare-seeking behavior for treatment of urinary symptoms: The Ohnsted County Study of Urinary Symptoms and Health Status Among Men. Arch Fam Med 2: 729-735, 1993. 10. Chute CG, Panser LA, Girman CJ, Oesterling JE, Guess HA, Jacobsen SJ, and Lieber MM: The prevalence of prostatism: A population-based survey of urinary symptoms. J Urol 150: 85-89.1993. 11. Kurland LT, Elveback LR, and Nobrega FT: Population studies in Rochester and Ohnsted County, Minnesota, 1900-1968, in Kessler 11, and Levin ML: 2Ie Community as an Epidemiologic Laboratory Baltimore, The Johns Ho$kins University Press, 1970, pp 47-70. 12. Epstein RS, Deverka PA, Chute CG, Panser LA, Oesterling JE, Lieber MM, Schwartz SW, and Patrick D: Validation of a new quality of life questionnaire for benign prostatic hyperplasia. J Clin Epidemio145: 1431-1445, 1992. 13. Barry MJ, Fowler FJ Jr, O’Leary MP, Bruskewitz RC, Holtgrewe HL, Mebust WK, and Cockett ATK: The American Urological Association symptom index for benign prostatic hyperplasia. The Measurement Committee of the American Urological Association. J Urol 148: 1549-1557, discussion 1564, 1992. 14. Dupuy HJ: The Psychological General Well-Being (PGWB) Index, in Wenger NK, Mattson ME, Furberg CD, and Elinson J (Eds): Assessment of Quality of Lije in Clinical Trials of Cardiovascular Therapies, New York, LeJacq, 1984, pp 170-183. 15. Levkoff SE, Cleary PD, and Wetle T: Differences in determinants of physician use between aged and middle-aged persons. Med Care 25: 1148-1160, 1987. 16. Klonoff EA, and Landrine H: Sex roles, occupational roles, and symptom-reporting: a test of competing hypotheses on sex differences. J Behav Med 15: 355-364, 1992. 17. Levkoff SE, Cleary PD, Wetle T, Besdine RW: Illness behavior in the aged. Implications for clinicians. J Am Geriatr Sot 36: 622-629,1988. 18. Keller ML, Leventhal H, Prohaska TR, and Leventhal EA: Beliefs about aging and illness in a community sample. Res Nurs Health 12: 247-255, 1989.
671
CHRISTOPHER G. CHUTE, M.D., DRPH. JOSEPH E. OESTERLING, M.D. MICHAEL M. LIEBER, M.D.
From the Section of Clinical Epidemiology and the Department of Urology, Mayo Clinic and Foundation, Rochester, Minnesota, and the Department of Epidemiology, Merck Research Laboratories, Blue Bell, Pennsylvania.
ABSTRACT-The objective of this study was to assess the association between frequency and bother of urinary symptoms in a population-based cohort of men and to identify psychosocial factors that are related to reporting heightened or subdued bother relative to symptom frequency. The survey was conducted among men aged forty to seventy-nine years in Olmsted County, Minnesota, the baseline component of a prospective cohort study. Men were queried about the frequency of urinary symptom occurrence and the perceived bother associated with the symptoms. A regression analysis of American Urologic Association (AUA) bother scores on AUA frequency scores demonstrated a tight correspondence (r2 = 0.71). Men with bother scores greater than predicted from their frequency scores were more likely to have sought health care for their urinary symptoms than men whose bother was close to predicted (14 versus 5 percent, respectively). These men with heightened bother were older, poorer, more anxious, and had lower general psychologic well-being scores than the men whose bother was similar to that expected from their reported frequency. Men whose bother was lower than would be expected were less likely to have sought health care for urinary symptoms in the past year (3%) but were of similar age and socioeconomic status as compared with men whose bother was close to expected. These men, however, were more depressed than men whose bother was commensurate with reported frequency. While the men who reported greater bother than expected from their symptom frequency were more likely to have sought medical care for urinary symptoms in the past year, it is not clear whether this greater healthcare-seeking behavior is because bother captures an additional component of urologic disease or is a manifestation of psychosocial differences.
Benign prostatic hyperplasia (BPH) is a common disease of men that is creating an increasing demand on the health care system.l In the United States, it has been estimated that 1.7 million physician office visits per year are related to BPH2 and, in 1989, approximately 340,000 men under*Tkis study was funded, in part, by research grantsfrom Merck Reseanh laboratories as part of the BPH Natural History Study Group and the National Institutes of Health (AR30582). Submitted: March 26, 1993, accepted (with revisions): June 1, 1993
UROLOGY /
DECEMBER 1993
/ VOLUME42, NUMBER6
went a transurethral prostatectomy3 Results from the Normative Aging Study have suggested that in the United States a forty-year-old man has a 30 percent chance of undergoing prostatectomy for BPH in the following forty years of life.4 The probability of treatment for BPH is likely to increase as more men may seek newly available medical therapy for their symptoms in order to avoid surgical management. While BPH is relatively common among older men, individual perceptions about the severity of
663
disease may vary. Previous research has noted that age-adjusted rates of prostatectomy for BPH vary considerably both between and within countries5-a stimulating the search for factors that might explain this tremendous small-area variation. Consequently, the focus of much of the health services research in urology has been on regional differences in practice patterns. Patterns in patient perceptions of urinary symptoms and the influence of these perceptions on the decision to seek care, however, have received much less attention. Nonetheless, treatment guidelines for BPH are currently being developed that take into account the frequency with which urinary symptoms occur. While it could be argued that the frequency of occurrence better defines the presence of disease, a recent study from this group suggested that the perceived bother from urinary symptoms is, in fact, a stronger predictor of health-care-seeking behavior for urinary dysfunction than is the frequency of symptom occurrence.9 This finding that bother from symptoms is more closely linked to health-care-seeking behavior than frequency seems straightforward, but little is known about why some men are bothered more (or less) by their urinary symptoms compared with the frequency with which symptoms occur. The Olmsted County Study of Urinary Symptoms and Health Status Among Men was designed, in part, to aid our understanding of the relationship between frequency and bother of urinary symptoms. During the baseline component, men from this population-based sample were queried about the frequency with which specific urinary symptoms occur as well as the perceived bother from these same symptoms. In this report, we describe the association between the frequency and bother of urinary symptoms. Further, we assess the relationship of several socioeconomic and psychometric characteristics with bother that is either subdued or heightened relative to reported frequency METHODS The Olmsted County Study of Urinary Symptoms and Health Status Among Men is a population-based, prospective cohort study initiated in December 1989 to study the natural history of benign prostatic hyperplasia. Many of the details of this study have been published previously.9v10 Briefly, this study was designed to measure the age-specific prevalence of urinary symptoms in a community-based sample of men and to follow these men over time for the onset as well as pro-
664
gression or regression of disease. Baseline symptom data were collected by questionnaire, and all study participants were asked to void into a standard urometer to determine the peak urinary flow rate. In addition, a probability sample of study participants was invited to a detailed urologic examination, which included a digital rectal examination and transrectal ultrasonography to estimate prostate size. The study protocol was reviewed and approved by the Mayo Institutional Review Board. Potential subjects included men aged forty to seventy-nine years who were residents of Olmsted County, Minnesota. The resources of the Rochester Epidemiology Project” were used to establish a sampling frame of Olmsted County residents. Men aged forty to seventy-nine years on January 1, 1990 were randomly selected within five-year age groups and two geographic strata (city of Roth es t er versus the balance of Olmsted County) at a 14 percent sampling fraction. After screening a potential subject’s medical history for exclusion and eligibility criteria,lO he was contacted and his participation in the study solicited. The exclusion criteria were established to remove men who had previously received surgical treatment for genitourinary conditions or had neurologic conditions thought to interfere with normal voiding. A small group of men with billing disputes or who were incarcerated in the Federal Medical Prison was also eliminated. Each participant was administered a structured, in-person interview eliciting information about current daily medications and family history of urologic disease by a trained field research assistant. After the interview, each man was asked to void into a standard urometer in privacy Finally, each man was administered a previously validated questionnaire12 while the research assistant remained in the home. This self-administered questionnaire included 134 elements about severity of and bother from urinary symptoms, health status, and several sociodemographic characteristics. There were twelve questions that elicited information about the frequency of specific urinary symptoms during the past month and an additional twelve questions queried the perceived bother of these same symptoms. Symptom frequency questions were measured on a 7-point scale from 0 (I do not have symptoms) to 6 (always); symptom bother questions were asked on a similar 7-point scale from 0 (no symptom) to 6 (extremely bothersome). Subjects were also queried about the degree of interference due to urinary problems for seven daily
UROLOGY
/ DECEMBER 1993 / VOLUME42, NLJMBER 6
activities including: drinking fluids before travel, drinking fluids before going to bed, driving for two hours without stopping, getting enough sleep at night, going places that may not have a toilet, playing sports outdoors, and going to movies, shows, church, etc. Responses to each of these activities were placed on a 5-point scale from 1 (none of the time) to 5 (all of the time). A composite score for symptom frequency was calculated to approximate the AUA symptom index for BPH13 by resealing our data in consultation with the AUA index developers as previously reported.9 Similarly, a composite score was calculated for bother to approximate the AUA bother score that is in development. In addition to these questions about urinary symptoms, the questionnaire included a subset of the General Psychologic Well-being Scale. 12s14 This scale consists of ten elements which are summed for an indication of general psychologic well-being. These elements also can be grouped into five subscales: anxiety, depression, positive well-being, self control, and vitality All ten elements are measured on a 6-point semantic differentiation scale from 1 to 6 with lower numbers representing more-impaired well-being. Finally, men were queried about several sociodemographic characteristics, including age, education, income, marital status, and employment status. METHODSOF ANALYSIS
For each of the twelve symptoms, a Spear-man rank-order correlation coefficient was calculated to measure the association between symptom bother and symptom frequency In order to classify a man according to the agreement between his frequency of urinary symptoms and their perceived bother, we compared a man’s AUA bother score with his AUA frequency score. Because these two scores are based on different scales (O-27 and O-34 for bother and frequency, respectively), we constructed a least squares regression model of AUA bother score on AUA frequency score. The coefficient of determination (r2> calculated from this model describes the amount of variance in bother score that is explained by variance in the frequency score. Further, the slope and intercept derived from the regression model predict an AUA bother score from the reported AUA frequency score, taking into account the scale differences and any overall downplaying of symptoms. The comparison of a man’s reported AUA bother score to the bother score predicted by the regression equation gives an indication of the
UROLOGY
/ DECEMBER1993
/ VOLUME42, NUMBER6
agreement between the frequency of urinary symptoms and their perceived bother. While the degree of departure from the line with which to label men as having heightened or subdued bother is somewhat arbitrary, for this study we consider a man whose observed bother score is more than 1 standard deviation (root mean square error) greater than his predicted bother score to have a heightened bother; whereas a man whose observed bother score was at least 1 standard deviation less than the predicted bother score is considered to have subdued bother. Men whose bother scores are within one standard deviation of their predicted bother score are considered to be concordant for symptom frequency and symptom bother. Similar cutpoints at 1.5 standard deviations led to similar results and only the former are presented. Based on the aforementioned classification, the distribution of demographic and psychosocial characteristics was compared among the three groups of men. A chi-square test of homogeneity was used to assess differences in proportions of factors among the three groups. Logistic regression models were constructed to simultaneously adjust for these demographic and psychosocial factors in predicting discordance between bother and frequency of symptoms. Models were constructed separately for men with heightened bother compared with men whose bother was close to predicted, and for men with subdued bother compared with men whose bother was close to predicted. Both backward- and forwardstepped algorithms were used to select the best set of predictor variables for each of the two models. Odds ratios were derived from the logistic regression models by exponentiating the estimated regression coefficients. For all statistical tests, a nominal p-value of co.05 was considered statistically significant. It should be recognized, however, that these p-values (and confidence intervals) do not take into account the multiple comparisons. All analyses were conducted using the SAS statistical packages (SAS Institute, Cary, NC). RESULTS Overall, the majority of men reported bother scores that were within one point of frequency scores for each of the twelve symptoms queried (Table I). Men were most concordant for dysuria, with 98 percent reporting a bother score within one point of frequency score, and least concordant for nocturia, with 85 percent of men being concordant. Of the men who were discordant, proportionately more reported bother scores at least 1 665
TABLE I.
Bother score relative to frequency for individual urinary symptoms (Olmsted County Study of Urinary Symptoms and Health Status Among Men) Bother Relative to Frequency’ Greater Bother n (4%)
Lower Bother n (%)
Similar n (%)
spearman Correlation
wet clothes
101 (4.8)
49 (2.3)
1949 (92.0)
0.80
Urgency
56 (2.6)
160 (7.6)
1900 (89.7)
0.73
Dribbling
42 (2.0)
202 (9.5)
1872 (88.3)
0.76
Repeat within 10 minutes
38 (1.8)
56 (2.6)
2022 (95.4)
0.05
Nocturia
36 (1.7)
252 (11.9)
1811 (85.5)
0.75
Incomplete empty
27 (1.3)
86 (4.1)
2003 (94.5)
0.71
Hesitancy
27 (1.3)
81 (3.8)
2007 (94.7)
0.71
Weak stream
26 (1.2)
254 (12.0)
1836 (86.6)
0.68
stop/start
25 (1.2)
151 (7.1)
1940 (91.6)
0.69
Daily frequency
24 (1.1)
265 (12.5)
1827 (86.2)
0.65
Dysuria
14 (0.7)
17 (0.8)
2085 (98.4)
0.66
Strain
15 (0.7)
71 (3.4)
2030 (95.8)
0.67
Symptomspresentedin otder ojdectraring magnitudeofproportion with heightened symptomawamness. Row totals may differdue to single item non-response. *GreaterBother-Bother scan at Ieast two greater thanjrequencyscan; Lower Bother-Bother score at least two less thanjrequencyscore;Similar-Bother scwe within one ojjrequencyscow
30:
AlJAmFIGURE 1. AUA bother score by AUA frequency score: Olmsted County Study of Urinary Symptoms and Health Status Among Men. Lines represent least-squares regression and -+1 standard deviation about the least-squares regression line. Bubble size is proportionate to number of men represented by the point.
666
UROLOGY
/ DECEMBER1993
I VOLUME42, NUMBER6
TABLE II.
Demographic and socioeconomic characteristics of men by concordance of observed and predicted urinary symptom bother [Olmsted County Study of Urinary Symptoms and Health Status Among Men) Heightened Bother’ (n=242) n (96)
Subdued Bother’ (n=266) n (4%)
Concordant (n= 1608) n (%)
~65
157 (65)
208 (78)
1279 (80)
265
85 (35)
58 (22)
329 (20)
44 (18)
25 (10)
154 (10)
High School Graduate
61 (26)
51 (20)
340 (21)
> High School Graduate
133 (56)
184 (71)
1099 (69)
<25
79 (34)
49 (20)
309 (20)
2545
73 (31)
75 (30)
475 (31)
245
83 (35)
126 (50)
751 (49)
Full-time
125 (48)
170 (64)
1089 (68)
115 (52)
95 (36)
512 (32)
Married
210 (87)
247 (93)
1436 (90)
Other
31 (13)
19 (7)
167 (10)
Characteristic Age, yrt
Education’
Income, $ (1,000’s)*
Current JQnploymentt
Marital Status
Health-CarAeeking
Behavioti
YeS
34 (14)
8 (3)
86 (5)
No
205 (86)
253 (97)
1505 (95)
Columntotals may diner due to single item non-response. *Heightenedbother;ObservedALJAbotherscore more than one standarddeviation above pwdicted bother scorefrom regressionof AUA bother score on AlJAfquency score. tSubdwd bother; ObservedAUA bother SCORmore than one standarddeviation below predictedbother score from regressionof AUA botherSCOH on AUAfrequcncyscore. +x2 test, p < 0.001. §x’ test, p = 0.05.
lower than frequency score for every symptom except for wet clothing. Overall, 259 men reported a greater bother than frequency score for at least one symptom, 930 reported a lower bother than frequency score for at least one symptom, and 1,030 men reported bother scores within 1 point for all twelve symptoms (116 men reported lower bother for at least one symptom and greater bother for another symptom). Each man’s AUA bother score is plotted against his AUA frequency score in Figure 1. AUA bother scores were very closely related to AUA frequency
UROLOGY
I DECEMBER1993
! VOLUME42, NUMBER6
scores, with a coefficient of determination of 0.71. Based on the regression analysis, 1,608 men (76%) had bother scores within one standard deviation of their bother score predicted on the basis of AUA frequency score (between upper and lower lines, Fig. 1) Two hundred sixty-six men (13%) reported subdued bother, with bother scores more than one standard deviation below their predicted bother score; 242 men (11%) reported heightened bother, with bother scores more than one standard deviation above their predicted bother score.
667
TABLE III.
Psychologic well-being of men by concordance of observed and predicted urinary symptom bother [Olmsted County Study of Urinary Symptoms and Health Status Among Men) Scale
Heightened Bother’ SubduedBother’ (n=242) (n=266) n (%) n (W
ConcordantBother (n= 1608) n (W
General Well-Bei& 240
208 (86)
237 (90)
1512 (94)
<40
34 (14)
27 (10)
93 (6)
r 10
174 (82)
202 (77)
1321 (82)
< 10
68 (18)
60 (23)
283 (18)
2 10
239 (99)
257 (98)
1586 (99)
< 10
3 (1)
5 (2)
14 (1)
r5
198 (82)
232 (88)
1432 (89)
<5
44 (14)
33 (12)
173 (11)
210
214 (88)
239 (91)
1514 (95)
< 10
28 (12)
24 (9)
87 (5)
210
194 (80)
222 (84)
1445 (90)
< 10
48 (20)
42 (16)
160 (10)
Anxiety
Positivity
Self-Control*
Vitality*
Depression*
For all scores, a higher score represents betterfunction. Column totals may di@ due to single item non-response. ‘Heightened bother: Observed AUA bother score more than one standard deviation above predicted bother scoreJrom regression of AUA bother score on AUA Jrequency score. tsubdued bother:Observed AUA bother scow more than one standanldeviation below predicted bother scoreJrom regression of AUA bother score on AUA frequency score. Q2 test of homogeneity, p < 0.01.
As would be expected, men with heightened bother were much more likely to have sought health care in the past year for their urinary symptoms than men with subdued bother or men whose observed and predicted bother scores were concordant (14% versus 3% and 5%, respectively). Men with heightened bother were older, less educated, less likely to be employed full-time, less likely to be married, and had lower incomes than men whose observed and predicted bother score were close to one-another (Table II). In contrast, men with subdued bother were similar to the men with concordant observed and predicted bother scores in nearly all these characteristics. Men with heightened bother also had lower psychologic well-being scores than men who were concordant
668
(Table III), overall and for each of the subscales. Men with subdued bother show similar trends, with lower psychologic well-being scores for overall general well-being, anxiety, depression, and vitality Scores for each of these scales, however, were intermediate to those for men with heightened bother and men who were concordant. After simultaneous adjustment for all factors using logistic regression, men with heightened bother remained more likely to be older, have lower income, have lower general psychologic well-being scores, and be more anxious than men who were concordant for urinary symptom bother and frequency (Table IV). For men with subdued bother, however, none of the demographic or socioeconomic factors remained significant in the
UROLOGY
/ DECEMBER1993
/ VOLUME42, NKMBER6
TABLE IV.
Multiple logistic regression of psychosocial characteristics with heightened or subdued bother (Olmsted County Study of Urinary Symptoms and Health Status Among Men) Heightened Bother’ Crude*
Adjusted’
Subdued Bother+ Crude
Adjusted
4% yr c65’
1.0
1.0
1.0
265
2.1 (1.6, 2.8)’
2.0 (1.4, 2.8)
1.1 (0.8, 1.5)
__
Education
2.4 (1.6, 3.4)
High school graduate
1.5 (1.1, 2.1)
__
1.0 (0.6, 1.5) 0.9 (0.6, 1.2)
1.0
__
1.0
<25
2.3 (1.7, 3.2)
1.7 (1.2, 2.5)
1.0 (0.7, 1.4)
25-45
1.4 (1.0, 1.9)
1.3 (0.9, 1.8)
1.0 (0.7, 1.3)
2451
1.0
1.0
1.0
1.0
__
> High school graduate1
__ __
Income, 1,OOOsof dollars/yr
__ __
Current Employment Status < Full-time1 Full-time
1.0 0.8 (0.6, 1.1)
0.5 (0.4, 0.7)
__ __
Marital Status 1.3 (0.8, 1.9)
__ __
0.7 (0.4, 1.1)
2.7 (1.8, 4.0)
1.9 (1.1, 3.3)
1.9 (1.2, 2.9)
1.0
1.0
1.0
No (2 10)’
1.0
1.0
1.0
Yes (C 10)
1.8 (1.3, 2.5)
1.7 (1.2, 2.5)
1.4 (1.0, 1.9)
Married1
1.0
Other
1.0
__ __
General Psychologic Well-Being C40 r40’
__ __
Anxiety
__ __
Positivity Yes (z-10)1 No (< 10)
1.0
1.0 1.4 (0.4, 5.0)
__
2.2 (0.8, 6.2)
1.0
__ __
1.2 (0.8, 1.8)
__ __
Self-Control Yes (z-5)’ No (
1.8 (1.3, 2.6)
1.0
__ __
Vitality Yes (ZlO)l No (< 10)
1.0
1.0
__ -_
2.3 (1.5, 3.6)
__
1.8 (1.1, 2.8)
1.0
__ __
1.0
1.0
1.7 (1.2, 2.5)
1.7 (1.1, 2.4)
Depression No (5: 10)’ Yes (
2.2 (1.6. 3.2)
*Heightenedbothn: Observed AUA botherscore more than one standarddeviationabove predictedbotherscorefrom regressionof AUA bother scan on AUAfreqwncy score. tSubduedbother:ObservedAUA botherscore more than one standurddeviationbelow predictedbotherscorefrom regressionof AUA bother score on AUAfrequencyscow. iBiva7iate association. §Adjusted_for other&ton in model (factorsp > 0.10 not includedin model). llRejerence category. POddsratio (95%confidenceinterval).
UROLOGY
/ DECEMBER1993
/ VOLUME42, NUMBER6
669
multivariable model; only depression was predictive of subdued bother relative to concordance. COMMENT In this community-based cohort of men, the frequency with which urinary symptoms were reported corresponded very closely to reported bother from urinary symptoms. There were, however, a number of men whose urinary symptoms were more bothersome than would be expected from their frequency of occurrence. These men with heightened bother tended to be older, poorer, and more anxious than men whose bother from urinary symptoms was comparable to the reported frequency. Similarly, there were a number of men who tended to downplay their urinary symptoms (with bother scores low relative to frequency scores). These men tended to be more depressed than men whose bother was comparable to their frequency of urinary symptoms, but were no different in terms of age, education, or income. The presence of heightened bother from urinary symptoms was reflected in health-care-seeking behavior; men with heightened bother were much more likely to have sought medical care in the past year for urinary symptoms than men whose bother was comparable or subdued relative to their urinary symptom frequency While the majority of men reported bother scores that were comparable to their frequency scores, there was, in general, a downward scoring of symptom bother relative to frequency. Obviously, a portion of this difference is attributable to differences in scales; the range of possible scores is smaller for bother than for frequency In addition, some portion of this downward scoring could be due to other factors. These could include a reluctance to admit to bother (a general stoicism) or, alternatively, some degree of acceptance. Unfortunately, we are unable to determine how much of the downward scoring is due to scale and how much is due to other factors. Regardless, as treatment recommendations are suggested to be based on frequency of urinary symptoms, it is reassuring that the amount of bother reported by most men in this community-based sample is in accord with their reported frequency of urinary symptoms. Among the men whose bother seems disproportionately greater than their reported frequency of urinary symptoms, there is a strong association with health-care-seeking behavior for urinary symptoms. The appropriateness of this heightened bother, however, is not easily discernible. The higher likelihood of seeking health care may be
due to clinical disease that is worse than indicated by frequency scores. That is, bother may add an additional dimension of disease that is not captured by frequency of symptom occurrence which should be considered in treatment decisions. Alternatively, heightened bother may not be related to worse clinical disease, but instead to other psychosocial factors.‘* The association with older age may be due to older men having more free time to contemplate symptoms than younger men. The younger men might also be more likely to experience social pressures to downplay bother.16 Further, older men may have been more likely to have grown weary of their symptoms and are at an age when acute changes might be more likely to be noticed and perceived as abnormal.r7 The association between heightened bother and anxiety is particularly interesting. If the heightened bother was due to anxiety, then perhaps men with heightened bother should be channeled into treatments that would address their anxiety and other psychologic aspects of their disease as opposed to surgical treatments. Unfortunately, these crosssectional data cannot distinguish the causal pathway In contrast to men with heightened bother, there was no association between age and subdued bother. This suggests that there is no additional acceptance of bother from symptoms’s with age beyond that accounted for by the regression of bother on frequency score. The association between depression and subdued bother suggests, however, an appealing causal hypothesis, that depressed men are reluctant to admit bother or have other, more pressing concerns. In the interpretation of these results, several potential limitations must be kept in mind. As mentioned, the cross-sectional nature of these data precludes the understanding of the temporal relationships between onset of symptoms, progression or regression of symptoms, and the association with onset of bother. The longitudinal follow-up of this cohort, however, should provide us with this information in the future. The exclusion of men who had already been surgically treated for their urinary symptoms may have selected out the most severely diseased men from the cohort. If these excluded men were more likely to have heightened bother, the association between heightened bother and health-care-seeking behavior in this study was most likely an underestimate of the true association. The 55 percent response rate must also be considered. If men who refused to participate were systematically more likely to have sought health care in the past year and were
UROLOGY
/ DECEMBER 1993 / VOLUME42, NUMBER6
more likely to report similar frequency and bother scores, our results might be biased. Further, respondents may differ from non-respondents systematically in the distribution of psychosocial parameters, providing the chance for further bias. These complex interactions, however, would probably not be strong enough to explain our findings. Finally, it must be recognized that multiple statistical tests and estimates have been calculated. Consequently, true p-values are greater than the nominal values presented and 95 percent confidence are wider than shown. The point estimates of the associations, however, remain unchanged. In conclusion, we found that while most men report a degree of bother associated with urinary symptoms that is in accord with the frequency of the symptoms, there are men in the community whose symptoms bother them more than the frequency with which their symptoms occur would suggest. These men with heightened bother appear to be more prone to have sought medical care for urinary symptoms in the past year, but it is not clear that this health-care-seeking behavior was inappropriate. As we follow these men over time, we should be able to sort out the appropriateness of this increased health-care-seeking behavior and develop a better understanding of BPH as a disease entity. Steven J. Jacobsen, M.D. DepartmentofHealth Sciences Research Mayo Clinic 200 First Street, SW Rochestel; Minnesota 55905 ACKNOWLEDGMENT. We thank the Ohnsted County Study of Urinary Symptoms and Health Status Among Men research staff, including Rebecca Nelson, Roger Mueller, and Charles Wilson for their assistance in the conduct of this study. In addition, we sincerely thank Ms. Sondra Buehler for her adept preparation of this manuscript. Finally, we wish to express our sincere appreciation of the enthusiasm of all members of the Ohnsted County Study of Urinary Symptoms and Health Status Among Men, without whose cooperation this study wouldnot be possible. , REFERENCES 1. Guess HA: Benign prostatic hyperplasia: antecedents and natural history. Epidemiol Rev 14: 131-153, 1992. 2. National Center for Health Statistics: United States Life Tables. U.S. Decennial Life Tables for 1979-81. Vol. 1, No.1. DHHS Pub. No. (PHS) 85-1150-1, Washington, D.C., U.S. Government Printing Office, 1985. 3. Graves EJ: Detailed Diagnoses and Procedures, National Hospital Discharge Survey, 1989. Vital Health Statistics 13( lOf3), Hyattsville, MD, National Center for Health Statistics, 199 1.
UROLOGY
/ DECEMBER 1993 / VOLUME42, NLIMBER6
4. Glynn RJ, Campion Ew, Bouchard GR, and Silbert JE: The development of benign prostatic hyperplasia among volunteers in the Normative Aging Study. Am J Epidemiol 121: 78-90,1985. 5. Wennberg JE, Mulley AG Jr, Hanley D, Timothy RP, Fowler FJ Jr, Roos NP. Barry MJ, McPherson K, Greenberg ER, Soule D, et al.: An assessment of prostatectomy for benign urinary tract obstruction. Geographic variations and the evaluation of medical care outcomes. JAMA 259: 3027-3030, 1988. 6. Wennberg J, and Gittelsohn A: Small area variations in health care delivery, Science 182: 1102-l 108, 1973. 7. McPherson K, Wennberg JE, Hovind OB, and Clifford P: Small-area variations in the use of common surgical procedures: an international comparison of New England, England, and Norway. N Engl J Med 307: 1310-1314,1982. 8. Stephenson Wp, Chute CG, Guess HA, Schwartt S, and Lieber M: Incidence and outcome of surgery for benign prostatic hyperplasia among residents of Rochester, Minnesota: 1980-87. A population-based study. Urology (Suppl 1) 38: 3242,199l. 9. Jacobsen SJ, Guess HA, Panser L, Girman CJ, Chute CG, Oesterling JE, and Lieber MM: A population-based study of healthcare-seeking behavior for treatment of urinary symptoms: The Ohnsted County Study of Urinary Symptoms and Health Status Among Men. Arch Fam Med 2: 729-735, 1993. 10. Chute CG, Panser LA, Girman CJ, Oesterling JE, Guess HA, Jacobsen SJ, and Lieber MM: The prevalence of prostatism: A population-based survey of urinary symptoms. J Urol 150: 85-89.1993. 11. Kurland LT, Elveback LR, and Nobrega FT: Population studies in Rochester and Ohnsted County, Minnesota, 1900-1968, in Kessler 11, and Levin ML: 2Ie Community as an Epidemiologic Laboratory Baltimore, The Johns Ho$kins University Press, 1970, pp 47-70. 12. Epstein RS, Deverka PA, Chute CG, Panser LA, Oesterling JE, Lieber MM, Schwartz SW, and Patrick D: Validation of a new quality of life questionnaire for benign prostatic hyperplasia. J Clin Epidemio145: 1431-1445, 1992. 13. Barry MJ, Fowler FJ Jr, O’Leary MP, Bruskewitz RC, Holtgrewe HL, Mebust WK, and Cockett ATK: The American Urological Association symptom index for benign prostatic hyperplasia. The Measurement Committee of the American Urological Association. J Urol 148: 1549-1557, discussion 1564, 1992. 14. Dupuy HJ: The Psychological General Well-Being (PGWB) Index, in Wenger NK, Mattson ME, Furberg CD, and Elinson J (Eds): Assessment of Quality of Lije in Clinical Trials of Cardiovascular Therapies, New York, LeJacq, 1984, pp 170-183. 15. Levkoff SE, Cleary PD, and Wetle T: Differences in determinants of physician use between aged and middle-aged persons. Med Care 25: 1148-1160, 1987. 16. Klonoff EA, and Landrine H: Sex roles, occupational roles, and symptom-reporting: a test of competing hypotheses on sex differences. J Behav Med 15: 355-364, 1992. 17. Levkoff SE, Cleary PD, Wetle T, Besdine RW: Illness behavior in the aged. Implications for clinicians. J Am Geriatr Sot 36: 622-629,1988. 18. Keller ML, Leventhal H, Prohaska TR, and Leventhal EA: Beliefs about aging and illness in a community sample. Res Nurs Health 12: 247-255, 1989.
671