Health problems associated with lower urinary tract symptoms in older women

Abstract

Objective. To identify health problems associated with lower urinary tract symptoms (LUTS) in an older female population. Design. Population-based cross-sectional survey. Setting. Third wave for the Tampere Longitudinal Study on Ageing (TamELSA) in 1999–2000. Subjects. A total of 203 women aged 70 years and over. Main outcome measures. LUTS categorized as urge symptoms only, voiding symptoms only, and urge and voiding symptoms combined. The health indicators were polypharmacy (> 3 medications), use of sleeping medication, constipation, treated urinary tract infections (UTIs), and neurological, cardiovascular, and musculoskeletal diseases. Results. Of the respondents 16% reported urge symptoms, 22% voiding symptoms, and 26% combined symptoms. In the multinomial logistic regression models, constipation and musculoskeletal diseases were significantly associated with voiding symptoms (OR 4.33; 95% CI 1.62–11.57 and OR 5.25; 95% CI 1.94–14.18, respectively) and with combined symptoms (OR 4.33; 95% CI 1.67–11.21 and OR 2.84; 95% CI 1.17–6.89, respectively). UTIs (OR 2.78; 95% CI 1.09–7.08) were associated with combined symptoms. Conclusions. Voiding symptoms comprise a significant part of lower urinary tract symptomalogy in older women. Although constipation, musculoskeletal diseases, and UTIs are the main health problems associated with LUTS, other concomitant diseases and medications also need to be assessed in older female patients with LUTS.

• Aged
• female gender
• voiding dysfunction

Older people comprise an increasing group of patients in the healthcare systems of ageing societies. Urinary problems, often as one element in comorbidity, are common in this age group, a circumstance that calls for a specific attention on the part of general practitioners.

Older patients using primary healthcare in either outpatient or institutional settings may present a variety of lower urinary tract symptoms (LUTS) related to both storage and voiding phases of the micturition cycle, both singly and in combinations. In addition to stress incontinence, urgency and urge incontinence, the key symptoms of an overactive bladder, are known to be especially common in older women [1]. Voiding symptoms are traditionally considered to be frequent among older men with bladder outlet obstruction caused by benign prostatic enlargement. According to well-established data, older women also frequently report such symptoms [2], [3]. Even though generally less bothersome than storage symptoms, voiding symptoms are significant in that they might be suggestive of voiding difficulties. Non-sex-specific changes in the ageing detrusor, for example an underactive detrusor, have been suggested as one explanation for this [4], but other possible contributing factors remain by and large unknown.

While several medical conditions have been found to be associated with urinary incontinence, especially the urge type in older populations [5], [6], much less is known of health problems associated with voiding symptoms. Non-urological diseases such as constipation, faecal incontinence, arthritis and neurological diseases are reported to be associated with LUTS in middle-aged and elderly men [7]. As far as we know, no reports are so far available on the association of various health conditions with LUTS other than urinary incontinence in older women.

The aim here was to explore health problems associated with LUTS with a specific emphasis on voiding symptoms in a randomly selected female population aged 70 years and over.

Material and methods

Study population

The data came from the Tampere Longitudinal Study on Ageing (TamELSA), a prospective survey covering both home-dwelling and institutionalized subjects. The survey was initiated in 1979 and the original study design has been described elsewhere [8]. As a part of the third wave of the TamELSA in 1999–2000, 227 women aged 70 years and older were interviewed. The response rate was 93%. Owing to the original stratified sampling method of the survey, a specific weighting method was used to allow generalization of the figures to the basic population of Tampere at the time of the survey. The weighting process has been described in greater detail elsewhere [3]. Due to missing data 203 women were analysed.

Variables

In the third wave of the TamELSA, urinary symptoms were enquired after in greater detail than in the earlier stages of the study. At the opening of the questionnaire the following introduction was given: “Urinary symptoms are known to be common among elderly people. I am now going to ask you some questions on them.”

In order to define voiding symptoms, the following question was asked: “Has your urinary stream become weaker than before or is there an intermittent stream?”

For measurement of urge symptoms respondents were asked whether they had trouble in getting to the lavatory in time or whether the urge to urinate became so strong that there was urinary leakage. For the purpose of the present study, the urinary symptoms were divided into three categories: reporting urge symptoms only, reporting voiding symptoms only, and reporting both urge and voiding symptoms.

Constipation was defined as whether or not reporting the symptom regardless of frequency (occasionally, often, all the time). Urinary tract infections (UTIs) were defined as reporting at least one infection treated with antibiotics during the previous two years. Polypharmacy was defined as taking at least four medications prescribed by a physician during the three previous months. As information on medication was missing for 11%, a separate category “missing” was added to this variable. Use of sleeping medication was defined as using or not using a medicine to help fall asleep.

The participants were asked to report any diseases or injuries diagnosed by a doctor. Neurological, musculoskeletal, and cardiovascular diseases were defined as conditions reported according to the ICD-9.

Statistical analyses

Cross-tabulations were used to describe the weighted percentage distribution of health problems according to urinary symptom groups. The differences between the groups were compared using the Pearson chi-squared test.

Multinomial logistic regression models with odds ratios (OR) and 95% confidence intervals (CI) were used to examine the association of health problems with urge symptoms only, voiding symptoms only, or with urge and voiding symptoms combined. In the analyses, those reporting no urinary symptoms were used as the reference group for each separate urinary symptom group. Age and all health problems were introduced into the model simultaneously. The urinary symptom categories constituted the outcome variables.

Statistical analyses were performed using SPSS for Windows, version 12.0. A p-value of less than 0.05 was considered statistically significant.

Results

Distribution of urinary symptoms and other health problems in the study population

Urge symptoms only were reported by 16% of the respondents, while the corresponding figure for voiding symptoms only was 22%. Altogether 26% of the study population reported urge and voiding symptoms simultaneously while 36% reported no LUTS (figures not shown in the tables). The most frequently reported health problems (>60%) were cardiovascular and musculoskeletal diseases (Table I).

Table I.  Distribution of health factors according to urinary symptoms in the study population (n = 203).

	Urinary symptoms
	No	Urge	Voiding	Both	All
	n = 74	n = 32	n = 45	n = 52	n = 203
	%	%	%	%	%	p
Age						0.446
70–79 years	70	72	73	60	69
At least 80 years	30	28	27	40	31
Polypharmacy						*0.021
0–3	49	45	44	36	45
At least 4	25	49	53	62	44
Missing	26	6	2	2	11
Use of sleeping medication						0.066
No	33	27	53	42	61
Yes	67	73	47	58	39
Treated urinary tract infections						0.004
No	85	79	67	58	74
Yes	15	21	33	42	26
Constipation						0.004
No	85	75	60	60	72
Yes	15	25	40	40	28
Neurological diseases						0.540
No	69	58	62	58	63
Yes	31	42	38	42	37
Cardiovascular diseases						0.066
No	43	34	29	21	33
Yes	57	66	71	79	67
Musculoskeletal diseases						<0.001
No	53	45	18	27	37
Yes	47	55	82	73	63

Differences between groups were tested by Pearson chi-squared test. *Without missing; with missing p < 0.001.

Association of urinary symptoms with health problems

According to the cross-tabulations, polypharmacy, treated urinary tract infections, constipation, and musculoskeletal diseases were significantly more frequent among women with urinary symptoms compared with women with no such symptoms (see Table I).

In the multinomial logistic regression models (Table II), each urinary symptom group was separately compared with respondents reporting no urinary symptoms. Constipation and musculoskeletal diseases were significantly associated with voiding symptoms only and with urge and voiding symptoms combined.

Table II.  Association of urinary symptoms with health indicators in the study population (n = 203).

	Urinary symptoms
	Urge (n = 32)	Voiding (n = 45)	Both (n = 52)
	OR (95% CI)	OR (95% CI)	OR (95% CI)
Age
70–79 years	1	1	1
At least 80 years	0.69 (0.25–1.86)	0.64 (0.25–1.66)	1.21 (0.51–2.87)
Polypharmacy
0–3 medicines	1	1	1
At least 4 medicines	2.56 (0.93–7.08)	1.78 (0.68–4.59)	2.44 (0.98–6.06)
Missing	0.19 (0.03–1.11)	0.08 (0.01–0.82)	0.18 (0.02–1.37)
Use of sleeping medication
No	1	1	1
Yes	0.47 (0.17–1.32)	1.65 (0.68–4.03)	0.87 (0.37–2.08)
Treated urinary tract infections
No	1	1	1
Yes	1.10 (0.35–3.41)	2.04 (0.75–5.54)	2.78 (1.09–7.08)
Constipation
No	1	1	1
Yes	2.19 (0.74–6.47)	4.33 (1.62–11.57)	4.33 (1.67–11.21)
Neurological diseases
No	1	1	1
Yes	1.69 (0.64–4.42)	1.75 (0.69–4.45)	1.59 (0.65–3.87)
Cardiovascular diseases
No	1	1	1
Yes	0.64 (0.22–1.84)	0.59 (0.21–1.64)	0.99 (0.35–2.79)
Musculoskeletal diseases
No	1	1	1
Yes	1.29 (0.52–3.24)	5.25 (1.94–14.18)	2.84 (1.17–6.89)

Multinomial logistic regression models were used, with results given as the odds ratios (OR) and 95% confidence intervals (CI). Reference group of all urinary symptoms is none (n = 74).

Treated UTIs were associated with urge and voiding symptoms combined when compared with respondents with no urinary symptoms. None of the health problems was associated with urge symptoms only.

Discussion

Our study assessed the association of health conditions with LUTS in a random older female population. Previously, Koskimäki and colleagues [7] examined the association of non-urological diseases with LUTS among 50-, 60-, and 70-year-old men in a general population. One of the strengths of the present study was that it allowed evaluation of the effect of a relevant variety of health factors on LUTS. Furthermore, our study focused on voiding symptoms, a urinary symptom group less frequently examined among older women. Moreover, we took into account the other major urinary symptom group, namely urge symptoms, in the analyses.

There are also evident limitations. First, the study population was relatively small, but our sample was clearly representative with a very high response rate. Urinary symptoms and other health problems were frequently reported by the respondents, this ensuring sufficient statistical power.

Second, no standardized questionnaires on urinary symptoms were used even though several have been developed. Even though tested among women [2], most questionnaires covering both storage and voiding symptoms have been originally designed for men with LUTS suggestive of benign prostatic enlargement. Nevertheless, the definitions we used for voiding symptoms and urge incontinence basically comply with the current definitions of lower urinary tract dysfunction introduced by the International Continence Society (ICS) [9]. A simple patient questionnaire to assist self-detection of an overactive bladder has recently been validated and found to correspond well with physicians’ diagnosis [10]. Information on all health problems was also based on self-reports, which may somewhat affect the accuracy of the data, a problem shared with most population-based studies [11].

The health conditions most strongly associated with LUTS in the present study were constipation and musculoskeletal diseases, which were associated with both voiding symptoms alone and voiding and urge symptoms combined. These findings are in fact in accord with those reported by Koskimäki and co-workers [7], who noted a significant association between LUTS and constipation and arthritis in men. They also found faecal incontinence and neurological diseases to be significantly associated with LUTS. LUTS were defined as a symptom index covering both storage and voiding LUTS. Given that faecal impaction is known to be associated with urinary retention in geriatric patients [12], [13], the association of constipation with voiding symptoms may suggest voiding difficulties. An association of constipation with urge incontinence has also been reported previously [5].

Diokno and colleagues [14] have noted an association between bladder-emptying symptoms and arthritis in men. Several causal pathways may explain the association of musculoskeletal diseases with voiding symptoms. Degenerative back diseases such as spinal stenosis might affect the complex innervation of the urinary bladder, which might in turn lead to impaired emptying of the bladder. As suggested by Koskimäki and colleagues [7], painkilling medication used to relieve musculoskeletal pain might also explain the association via a negative effect on the contractility of the urinary bladder.

The association of treated UTIs with voiding and urge symptoms combined is not straightforward. The basic limitation of the present study was that the indication for the treatment of UTI was not specified. However, there is a known tendency to overdiagnosis and overtreatment of UTIs among older patients, especially women [15]. Asymptomatic bacteriuria and LUTS similar to those possibly due to an acute UTI are both common among older people. Voiding difficulties and increased post-void residual volumes may nevertheless increase the risk of a symptomatic UTI [16].

Various drugs are known to have effects on lower urinary tract function. In particular, drugs with anticholinergic properties may negatively affect the contractility of the urinary bladder, leading to voiding difficulties and risk of urinary retention [17]. According to recent data, use of such psychotropic medications and sedatives is especially common in the older population [18]. In addition, drugs used for other indications, among them antimuscarinic agents indicated for the overactive bladder, may carry a risk of anticholinergic side-effects. Here, in the non-adjusted cross-tabulations, women using at least four medications were significantly more likely to report urinary symptoms than those using fewer medications. However, in the multinomial analyses there was no significant association between polypharmacy and any of the urinary symptom groups. One limitation of the present study was that medication groups other than sleeping medications were not specified.

Finally, it has to noted that even though voiding symptoms may suggest problems in bladder emptying, the correlation of self-reported symptoms with voiding difficulties has been poor in urodynamic examinations, especially in women [19], [20]. On the other hand, in a study by Stanton and associates [19], poor stream was the most common urinary symptom reported by women with voiding difficulties, and in another clinical study slow stream was the only LUTS associated with elevated residual volumes in hospitalized older women [13].

In conclusion, not only urinary incontinence but also voiding symptoms constitute a significant element in lower urinary tract symptomatology in older women and should be enquired after by general practitioners. The most important non-urological health problems associated with LUTS are constipation, musculoskeletal diseases, and UTIs. According to a recent study among community-dwelling elderly persons with urinary incontinence, incontinence-related symptoms, e.g. obstructive symptoms such as voiding problems, increased the likelihood of seeking help, especially in women [21]. Primary care physicians should apply a comprehensive approach when assessing their older female patients with LUTS. In addition to excluding urinary retention, most commonly by means of measuring the post-void residual volume, concomitant diseases, bowel function, and medications require attention.

Key Points

Urinary incontinence is common among older women while voiding symptoms are generally considered to be more prominent in older men.

• Voiding symptoms in our study constitute a significant proportion of lower urinary tract symptomalogy in older women.

• Constipation, musculoskeletal diseases, and urinary tract infections are associated with voiding symptoms with or without urge symptoms.

• Bowel function, concomitant diseases, and medications should be assessed in older women with lower urinary tract symptoms.

This work was financially supported by the Medical Research Fund of Tampere University Hospital.