You are here
Prevalence, Incidence, and Resolution of Nocturnal Polyuria in a Longitudinal Community-based Study in Older Men: The Krimpen Study
European Urology, Volume 63, Issue 1, 2013, pages 542 - 547
Nocturnal polyuria (NP) is common in older men and can lead to nocturia. However, no longitudinal data are available on the natural history of NP.
To determine prevalence, incidence, and resolution rates of NP.
Design, setting, and participants
A longitudinal, community-based study was conducted among 1688 men aged 50–78 yr in Krimpen aan den IJssel, The Netherlands (reference date: 1995), with planned follow-up rounds at 2, 4, and 6 yr.
Outcome measurements and statistical analysis
NP was determined with frequency–volume charts. Two definitions of NP were used: (1) a nocturnal urine production (NUP) of >90 ml/h (NUP90) and (2) the nocturnal voided volume plus first morning void being >33% of the 24-h voided volume (NUV33). Nocturia was defined as two or more voids per night. We determined the prevalence of NP at each study round. At first follow-up, we determined the incidence in men without baseline NP and the resolution in men with baseline NP. Prevalence of NP in men with or without nocturia was also determined.
Results and limitations
At baseline, the prevalence of NUP90 was 15.0% and increased to 21.7% after 6.5 yr, whereas the prevalence of NUV33 was 77.8% at baseline and 80.5% after 6.5 yr. At 2.1 yr of follow-up, the incidences of NUP90 and NUV33 were 13.6% and 60.3%, respectively, and the resolution rates were 57.0% and 17.8%, respectively. Because of this fluctuation in NP, no reliable long-term incidences could be calculated. At baseline, NUP90 was prevalent in 27.7% of men with nocturia and in 8.0% of those without nocturia. At baseline, NUV33 was prevalent in 91.9% of men with nocturia and in 70.1% of men without nocturia.
Due to the fluctuation of NP, it is advisable to first determine its chronicity and cause before starting treatment. Because of the high prevalence of NP in men without nocturia, NUV33 should be reconsidered as a discriminative definition of NP.
Keywords: Longitudinal community-based study, Frequency–volume charts, Nocturia, Nocturnal polyuria, Lower urinary tract symptoms, Aging male.
Nocturia is a common symptom in older men and is an important cause of sleep disruption. In 2002, the International Continence Society (ICS) defined nocturia as waking one or more times at night to void  and . Tikkinen et al. showed that nocturia becomes clinically relevant if a person has to get out of bed two or more times per night  .
Nocturia occurs when urine production at night is greater than the functional capacity of the bladder. An important cause of nocturia is nocturnal polyuria (NP). NP can be caused by conditions such as (subclinical) cardiovascular disease, sleep apnea syndrome, anxiety, excessive fluid intake, and the use of medication such as diuretics , , , and . Various definitions of NP have been proposed. The most widely accepted and used definition defines NP as a nocturnal urine production (NUP) ≥33% of the total 24-h urine production  . Blanker et al. proposed a urine production of ≥90 ml/h as the threshold for NP, as this would be the best predictor for nocturia in older men in a general population  .
Although NP seems to play an important role in the pathophysiology of nocturia, there are few data on its prevalence and incidence. Furthermore, based on longitudinal analysis, our group demonstrated that nocturia fluctuates over time in individual community-dwelling men  and ; therefore, we hypothesized that NP might also fluctuate over time.
To investigate this, we determined the prevalence and fluctuation of NP over time (according to two definitions) in community-dwelling men aged 50–78 yr using frequency–volume charts (FVCs). Furthermore, we explored the relationship between NP and clinically relevant nocturia as defined by nocturnal frequency that leads to a decrease in quality of life  .
2. Materials and methods
The Krimpen study is a longitudinal study on urogenital tract dysfunction and its impact on general health status. The design of this internal review board-approved study has been described previously  and . In short, all men aged 50–78 yr (reference date: June 1995; n = 3924] in the Dutch municipality Krimpen aan den IJssel were investigated. Exclusion criteria for participation were: previous transurethral or open prostatectomy, prostate or bladder cancer, neurogenic bladder disease, or negative advice from their primary care physician (PCP) based on poor health (eg, being bedridden). These exclusion criteria were chosen because we aimed to study the natural history in a community-dwelling population without previous surgical procedures of the lower urinary tract or existing neurogenic bladder disease.
At baseline, participants completed a 113-item questionnaire, including, among other items, the International Prostate Symptom Score and a medication list. Subsequently, participants visited the local PCP health center for medical examination. Next, appointments were made within 4 wk for further urologic measurements at the Erasmus Medical Centre Rotterdam urologic outpatient department. Participants were instructed to complete a FVC and bring it with them at this appointment. Subsequently, all data were entered in the database by a data manager.
The follow-up rounds were planned after average intervals of 2, 4, and 6 yr  . All baseline measurements, including the FVC, were repeated under the same conditions at the respective follow-up rounds. If the participant had not died or moved away, and no exclusion criteria were met, the PCPs were asked to send a reinvitation letter for a first, second, and third follow-up round. Men who had undergone lower urinary tract surgery during follow-up were excluded. Additionally, for the third follow-up round, all nonresponders of the previous rounds were reinvited.
Data collection for the baseline round started early 1996; the database was closed in April 2004 after completing data collection of the third follow-up.
2.2. Nocturnal urine production and voided volume determined from frequency–volume charts
On the FVC, participants reported each micturition in 1-h time units for 3 d. Additionally, they recorded the volume of each void on the third day. The recording of each day started at midnight (00:00) and ended at midnight (24:00), thus ensuring that the first morning void was always included  . The time of arising and bedtime were noted; fluid intake was not recorded. Retrospectively, this complies with the 2002 ICS definition of a FVC, which states that a FVC should at least record voided volumes for 24 h  .
We computed the hourly urine production according to the method of Van Mastrigt and Eijskoot  : Urine production was assumed to be constant between voids and hourly urine production was estimated as the volume of each micturition divided by the number of hours that had passed since the previous micturition. NUP was estimated as the mean hourly urine production (milliliters per hour) from 01:00 to 06:00, when >90% of the men were asleep.
The total nocturnal voided volume was determined by using participant-reported sleeping hours: Voided volumes between the hour of going to bed and the hour of arising plus the first morning void within 1 h of arising were summed for each participant.
2.3. Definitions of nocturnal polyuria
Two definitions of NP were used for the purpose of this analysis: nocturnal urine volume (NUV) >33% of 24-h total urine volume (NUV33)  and an average NUP >90 ml/h (NUP90)  . For each round, the prevalence of NP was determined for the different age strata based on the two definitions. Resolution was defined as NP at baseline but no NP at follow-up 1. In the subgroup of men who had no NP at baseline and also were investigated at follow-up 1, the incidence was determined by dividing the number of new cases of NP at follow-up 1 by the total number of participants without NP at baseline.
In the subgroup of men who had NP at baseline and also were investigated at follow-up 1, the resolution rate was determined by dividing the number of men who did not have NP at follow-up 1 by the number of men who did have NP at baseline.
We only determined the incidence at follow-up 1 because of the fluctuation of nocturia, which might imply a possible fluctuation of one of its most important causes (ie, NP)  .
We also determined how many men with and without nocturia had NP according to the two definitions (ie, NP prevalence according to nocturia status). Nocturia was defined as voiding two or more times per night.
2.4. Statistical analyses
All men were grouped into 5-yr age strata and their general characteristics were noted. For each stratum, the prevalence of NP was determined and compared throughout the follow-up rounds using the McNemar test. A p value <0.05 was considered statistically significant. Statistical analyses were performed with SPSS v.15.0 (IBM Corp., Armonk, NY, USA).
At baseline, 1597 men completed the FVC. Of these, 201 were excluded due to incomplete data, leaving 1396 men available for analyses. Due to participants who dropped out of the study, 958, 720, and 598 FVCs were available for analyses on NUP90 for follow-ups 1, 2, and 3, respectively. Because of missing sleeping hours, analyses on NUV33 were possible using baseline, follow-up 1, follow-up 2, and follow-up 3 data from 1067, 793, 393, 400 FVCs, respectively.
Characteristics of the baseline population are given in Table 1 . Median age was 60.7 (interquartile range: 55.8–66.1) yr. Most men (75.4%) had no (9.9%) or mild (65.5%) lower urinary tract symptoms and 27.1% had nocturia.
|Age, yr, median (IQR)||60.7 (55.8–66.1)|
|Age group, yr, no. (%)|
|IPSS score, median (IQR)||4 (1–7)|
|IPSS categories, no. (%)|
|None (0)||138 (9.9)|
|Mild (1–7)||914 (65.5)|
|Moderate (8–19)||304 (21.8)|
|Severe (20–35)||40 (2.9)|
|Prevalence, no. (%)|
|Nocturia *||27.1 (379)|
* Defined as two or more voids per night.
IPSS = International Prostate Symptom Score; IQR = interquartile range.
3.1. Noctural polyuria defined as urine production of >90 ml/h
The overall prevalence of NUP90 increased from 15.0% to 21.7% after 6.5 yr ( Table 2 ). Table 3 shows the prevalence of NUP90 in men with and without nocturia. NUP90 was prevalent in 27.7% of men with nocturia at baseline and increased to 55.9% at follow-up 3, and was prevalent in 8.0% of men without nocturia at baseline and in 19.8% at follow-up 3.
|Prevalence (percent) of NP >90 ml/h|
|Age group, yr||Baseline
n = 1396
at 2.1 yr
n = 958
at 4.2 yr
n = 720
at 6.5 yr
n = 598
|50–54||11.8 (8.0–15.5) *||14.7 (9.7–19.8)||19.3 (12.9–25.7)||19.3 (12.8–25.8)|
|55–59||12.2 (8.8–15.6)||12.1 (8.0–16.1)||15.9 (10.8–21.0)||18.3 (12.5–24.2)|
|60–64||12.5 (9.0–16.1)||21.6 (16.4–26.8)||15.3 (10.2–20.5)||26.4 (19.2–33.5) **|
|65–69||20.4 (15.5–25.3)||21.1 (15.0–27.3)||23.2 (15.7–30.7)||22.0 (13.7–30.3)|
|70–78||23.8 (16.8–30.8)||23.2 (14.8–31.7)||18.2 (7.7–28.7)||27.8 (12.4–43.2)|
|Total||15.0 (13.1–16.9)||17.8 (15.4–20.3)||17.9 (15.1–20.7)||21.7 (18.4–25.1) **|
|Prevalence (percent) of NP 33% of total 24-h voided volume|
|Age group, yr||n = 1067||n = 793||n = 393||n = 400|
|50–54||70.8 (64.5–77.1)||65.1 (57.5–72.8)||75.4 (64.6–86.1)||74.5 (65.7–83.3)|
|55–59||74.2 (67.0–79.4)||73.1 (67.0–79.2)||82.1 (74.9–89.4)||78.2 (70.3–86.0)|
|60–64||78.4 (73.3–83.5)||77.4 (71.7–83.1)||80.8 (73.1–88.5)||83.0 (75.8–90.3)|
|65–69||85.7 (80.9–90.5)||84.1 (78.1–90.2)||88.3 (81.0–95.7)||89.8 (81.9–97.8)|
|70–78||83.3 (76.7–89.9)||86.3 (78.5–94.0)||94.3 (86.2–100)||81.5 (65.8–97.1)|
|Total||77.8 (75.4–80.4)||76.0 (73.1–79.0)||83.0 (79.2–86.7)||80.5 (76.9–84.4)|
* Data given as % (95% confidence interval).
** McNemar test compared to baseline, p < 0.05.
NP = nocturnal polyuria.
|NP defined as a nocturnal urine production >90 ml/h|
|Age group, yr, %||Baseline Nocturia
n = 379
n = 697
n = 132
n = 675
n = 92
n = 330
n = 68
n = 343
|NP defined as >33% of the total 24-h voided volume|
|Age group, yr, %||n = 370||n = 689||n = 128||n = 656||n = 86||n = 306||n = 66||n = 372|
* Nocturia defined as two or more voids per night.
NP = nocturnal polyuria.
Table 4 shows the incidence and resolution rate after 2.1 yr. The overall incidence for NUP90 was 13.6% and the overall resolution rate was 57.0%. The incidence was highest in men aged 60–64 yr and lowest in the group aged 50–54 yr. Resolution was highest in men aged 55–59 yr and lowest in those aged 65–69 yr.
|NP defined as a nocturnal urine production >90 ml/h|
|Age group, yr||Incidence (n = 779)||Resolution (n = 135)|
|50–54||10.7 (5.8–15.6) *||60.0 (36.5–83.5)|
|55–59||11.3 (7.0–15.6)||78.3 (60.0–96.5)|
|60–64||17.8 (12.5–23.1)||52.9 (35.3–70.6)|
|65–69||12.8 (7.0–18.5)||43.8 (25.6–61.9)|
|70–78||16.7 (7.9–25.5)||57.7 (37.3–78.0)|
|Total||13.6 (11.2–16.0)||57.0 (48.6–65.5)|
|NP defined as >33% of the total 24-h voided volume|
|Age group, yr||Incidence (n = 131)||Resolution (n = 472)|
|50–54||58.6 (39.6–77.7)||25.0 (14.8–35.3)|
|55–59||70.7 (56.2–85.3)||24.8 (17.1–32.5)|
|60–64||48.6 (31.2–66.0)||14.9 (8.4–21.3)|
|65–69||66.7 (39.6–93.7)||12.1 (5.6–18.7)|
|70–78||54.6 (19.5–89.6)||9.1 (1.25–16.9)|
|Total||60.3 (51.8–68.8)||17.8 (14.3–21.3)|
* Data given as % (95% confidence interval).
NP = nocturnal polyuria.
3.2. Noctural polyuria defined as a nocturnal urine production >33% of the total 24-h urine production
The overall prevalence of NUV33 increased from 77.8% at baseline to 80.5% at follow-up 3 ( Table 2 ). Table 3 shows the prevalence of NUV33 in men with and without nocturia. NUV33 was prevalent in 91.9% of men with nocturia at baseline and in 93.9% at follow-up 3; it was also prevalent in 70.1% of men without nocturia at baseline and 78.0% at follow-up 3.
Table 4 shows the incidence and resolution rate after 2.1 yr. The overall incidence was 60.3%, and the overall resolution rate was 17.8%. The incidence was highest in men aged 65–69 yr and resolution was highest in those aged 50–54 yr.
This study shows that the prevalence of NP in community-dwelling older men is largely dependent on the definition used. NP is present in about 15% of older men in the open population when we apply a cut-off value of NUP >90 ml/h to indicate NP  . However, when using the more common definition of NP of >33% of the total 24-h voided volume, NP would be present in about 80% of older men. Resolution rates were high when using the NUP90 definition, whereas the incidence was very high when using the NUV33 definition.
Furthermore, NUP90 was prevalent in 27.7% of men with nocturia and in 8.0% of those without nocturia; NUV33 was prevalent in 91.9% of men with nocturia but also in 70.1% of those without nocturia. As NP is not the only cause of nocturia, no definition will give a perfect match between NP and nocturia.
Since, to our knowledge, this is the first report on prevalence and incidence of NP in an open population of older men, it is not possible to make comparisons with other studies. However, a recent study by Weiss et al. showed that in male nocturia patients, 90% had NUV33  . Koseoglu et al. found similar results in a Turkish population of benign prostatic hypertrophy (BPH) patients with nocturia  . Unfortunately, neither of those studies included men without nocturia for comparison. Our study shows that NUV33 is highly prevalent in both men with and men without nocturia. Asplund et al. found that 36.7% of elderly men (mean age: 73.0 yr) in Sweden have NP, but this was determined by an affirmative or negative response to the statement, “I am passing large amounts of urine at night,” and therefore was highly subjective  . In our opinion, that method is not sufficiently accurate to determine whether or not one has an increased NUP. In women aged >50 yr in the open population, Swithinbank et al. reported that NUV33 was present in 53% with a prevalence of >60% in women aged ≥70 yr; they also showed that there was no relation between NUV33 and nocturia  .The study by Weiss et al. also showed that NUV33 was less prevalent in women  .
Because NUV33 is highly prevalent in both men with and men without nocturia, we question the discriminative value of this definition for NP as a cause of nocturia. In our opinion, this is due to the fact that the NUV33 definition was first described for a group of nocturia patients  , but its specificity has never been tested in the open population. The proposed definition of NUP90, albeit only a modest predictor of nocturia, seems to be a more discriminative definition of NP  .
The high variability (ie, incidence and resolution) of NP might be explained, in part, by the fact that volumes were only recorded over a period of 24 h at each measurement. However, in a large epidemiologic study such as ours, a balance must be found between precision of the data and a response rate for the particular measurement that is as high as possible. In our study, 95% of the men who responded to the invitation to participate in this longitudinal study completed the FVC at baseline. Furthermore, the large number of participants may attenuate possible variability.
A limitation of our study is that we only used sleeping hours reported by participants for the determination of NUV33, even though this led to the exclusion of some men from the analysis. However, the characteristics of the men without data on sleeping hours did not differ from those who did report them  . Moreover, using fixed sleeping hours could lead to a miscalculation of the prevalence of NP due to the wide range in individual sleeping hours  .
Another possible limitation of our study is the fact that the rates of men lost to follow-up might create a risk of bias. However, previous analysis of nonresponders showed that there was no difference in characteristics of men who did or did not respond  . Furthermore, analysis showed that there was no difference in characteristics between men who did or did not reported their sleeping hours (data not shown).
The resolution of NP might be explained, in part, by medical treatment. However, none of our men received desmopressin and almost none stopped using diuretics. Furthermore, only a few men (3.1%) used alpha-blockers for BPH treatment, despite the finding that these medications have no significant effect on NP  . Therefore, we feel that medical treatment, if any, had a negligible impact on the resolution of NP. Another explanation for fluctuation in the prevalence of NP might lie in the multifactorial etiology , , and . For example, excessive fluid intake can easily be resolved without the need of a health care professional. Thus, we feel it is important to determine the cause of NP before starting treatment.
Although NP seems to be an important cause of nocturia, this study shows that it is often a transient phenomenon. Therefore, we feel that multiple FVCs should be completed by a nocturia patient over a period of time, and that lifestyle changes (eg, restricting fluid intake in the evening) should be implemented before the diagnosis NP can be justified and medical treatment started.
The proposed NP definition of NUP90  has a higher discriminative value than NUV33. However, more studies should apply the NUP90 definition to determine its generalizability. Moreover, since both NP definitions are only modest predictors of nocturia  , more attention should be paid to other possible predictors of nocturia, such as nocturnal bladder capacity or the interaction between NP and nocturnal bladder capacity.
Future studies should also focus on the etiology of NP to determine what causes resolving and continuous NP. Furthermore, future, randomized, controlled trials for possible treatment of NP should also take the fluctuation of NP into consideration when calculating sample size. Now that we have established incidence, prevalence, and fluctuation of NP, it will be possible to study the possible determinants and prognostic factors of incident NP. More importantly, the determinants of nocturia, of which NP is one, have to be analyzed. In particular, the possible identification and role of modifiable determinants might have an impact on the management of nocturia.
NP is highly prevalent in older, community-dwelling men but is often a transient phenomenon. Fluctuation in the prevalence of NP occurs regardless of the definition used and might be due to the multifactorial etiology of NP. Before starting treatment for NP as a cause of nocturia, it is advised to first determine the cause and chronicity. Furthermore, because of the high prevalence of NP in men with and without nocturia when using the NUV33 definition, this definition of NP does not discriminate between those two groups.
Author contributions: Boris van Doorn had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: van Doorn, Blanker, Kok, Bosch.
Acquisition of data: Kok, Blanker.
Analysis and interpretation of data: van Doorn, Blanker, Kok, Westers, Bosch.
Drafting of the manuscript: van Doorn.
Critical revision of the manuscript for important intellectual content: Blanker, Kok, Westers, Bosch.
Statistical analysis: van Doorn, Westers.
Obtaining funding: Bosch.
Administrative, technical, or material support: None.
Other (specify): None.
Financial disclosures: Boris van Doorn certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.
Funding/Support and role of the sponsor: This research is sponsored by an unrestricted educational grant by Ferring Pharmaceuticals.
-  P. Abrams, L. Cardozo, M. Fall, et al. The standardisation of terminology of lower urinary tract function: report from the Standardisation Sub-committee of the International Continence Society. Neurourol Urodyn. 2002;21:167-178 Crossref
-  P. Van Kerrebroeck, P. Abrams, D. Chaikin, et al. The standardisation of terminology in nocturia: report from the Standardisation Sub-committee of the International Continence Society. Neurourol Urodyn. 2002;21:179-183 Crossref
-  K.A.O. Tikkinen, T.M. Johnson II, T.L.J. Tammela, et al. Nocturia frequency, bother, and quality of life: how often is too often? A population-based study in Finland. Eur Urol. 2010;57:488-498 Crossref
-  R. Asplund. Nocturia, nocturnal polyuria, and sleep quality in the elderly. J Psychosom Res. 2004;56:517-525 Crossref
-  J.L. Bosch, J.P. Weiss. The prevalence and causes of nocturia. J Urol. 2010;184:440-446 Crossref
-  J.P. Weiss, J.G. Blaivas. Nocturnal polyuria versus overactive bladder in nocturia. Urology. 2002;60(Suppl 1):28-32 discussion 32 Crossref
-  J.N. Cornu, P. Abrams, C.R. Chapple, et al. A contemporary assessment of nocturia: definition, epidemiology, pathophysiology, and management – a systematic review and meta-analysis. Eur Urol. 2012;62:877-890 Crossref
-  M.H. Blanker, R.M. Bernsen, J.L. Bosch, et al. Relation between nocturnal voiding frequency and nocturnal urine production in older men: a population-based study. Urology. 2002;60:612-616 Crossref
-  J.T. Hakkinen, M. Hakama, R. Shiri, A. Auvinen, T.L. Tammela, J. Koskimaki. Incidence of nocturia in 50 to 80-year-old Finnish men. J Urol. 2006;176:2541-2545 discussion 2545 Crossref
-  B. Van Doorn, M.H. Blanker, E.T. Kok, P. Westers, J.L. Bosch. Once nocturia, always nocturia? Natural history of nocturia in older men based on frequency-volume charts: the Krimpen study. J Urol. 2011;186:1956-1961 Crossref
-  M.H. Blanker, A.M. Bohnen, F.P. Groeneveld, R.M. Bernsen, A. Prins, J.L. Ruud Bosch. Normal voiding patterns and determinants of increased diurnal and nocturnal voiding frequency in elderly men. J Urol. 2000;164:1201-1205
-  M.H. Blanker, F.P. Groeneveld, A. Prins, R.M. Bernsen, A.M. Bohnen, J.L. Bosch. Strong effects of definition and nonresponse bias on prevalence rates of clinical benign prostatic hyperplasia: the Krimpen study of male urogenital tract problems and general health status. BJU Int. 2000;85:665-671
-  E.T. Kok, F.P. Groeneveld, J.J. Busschbach, et al. Influence of coping styles on quality of life in men with new and increasing lower urinary tract symptoms. The Krimpen Study in community-dwelling men. Urol Int. 2007;79:226-230 Crossref
-  R. Van Mastrigt, F. Eijskoot. Analysis of voided urine volumes measured using a small electronic pocket balance. Scand J Urol Nephrol. 1996;30:257-263 Crossref
-  J.P. Weiss, P.E. van Kerrebroeck, B.M. Klein, J.P. Norgaard. Excessive nocturnal urine production is a major contributing factor to the etiology of nocturia. J Urol. 2011;186:1358-1363 Crossref
-  H. Koseoglu, G. Aslan, I. Ozdemir, A. Esen. Nocturnal polyuria in patients with lower urinary tract symptoms and response to alpha-blocker therapy. Urology. 2006;67:1188-1192 Crossref
-  R. Asplund. Hip fractures, nocturia, and nocturnal polyuria in the elderly. Arch Gerontol Geriatr. 2006;43:319-326 Crossref
-  L.V. Swithinbank, S. Vestey, P. Abrams. Nocturnal polyuria in community-dwelling women. BJU Int. 2004;93:523-527 Crossref
-  J.P. Weiss, J.G. Blaivas, D.S. Stember, M.M. Brooks. Nocturia in adults: etiology and classification. Neurourol Urodyn. 1998;17:467-472 Crossref
-  B. Van Doorn, E.T. Kok, M.H. Blanker, P. Westers, J.L.H.R. Bosch. Mortality in older men with nocturia. A 15-year followup of the Krimpen Study. J Urol. 2012;187:1727-1731 Crossref
-  M.H. Blanker, J. Prins, J.L. Bosch, et al. Loss to follow-up in a longitudinal study on urogenital tract symptoms in Dutch older men. Urol Int. 2005;75:30-37 Crossref
a University Medical Center Utrecht, Department of Urology, Utrecht, The Netherlands
b University of Groningen, University Medical Center Groningen, Department of General Practice, Groningen, The Netherlands
c University Medical Center Utrecht, Julius Center for Health Sciences and Primary Care, Utrecht, The Netherlands
© 2012 Published by Elsevier B.V.