Involvement of personal and professional relations among men bothered by lower urinary tract symptoms: a population-based cross-sectional study

This population-based, cross-sectional study is based on data from the Danish Symptom Cohort, a nationwide cohort comprising a random sample of 100,000 people aged 20 years or above who are representative of the adult Danish population.

All Danish citizens are registered with a unique personal identification number in the Danish Civil Registration System (CRS), which contains information on date of birth, gender etc. [17]. Invitees were randomly selected from the general population through the CRS and invited to participate in the survey by a letter explaining the purpose of the study. The letter included a unique login to a secure web page, which provided access to a comprehensive questionnaire. After 2 weeks a reminding letter was sent to non-respondents and after additional 2 weeks non-respondents were contacted by telephone and encouraged to participate. A telephone interview conducted by trained interviewers was offered to prevent the exclusion of people with no internet access. Data were collected from June to December 2012 and the methodological framework for developing and testing the questionnaire has been thoroughly described by Rasmussen et al. [18]. In this study, only male respondents were included.

The dataset for this study comprises men with bothersome LUTS, defined as being of moderate to extreme concern and/or with moderate to extreme influence on daily activities. This means that symptoms reported as being of no or slight concern and/or with no or slight influence of daily activity were not included in the analyses. For further details of the distribution of bothersome LUTS, see Rubach et al. [19]. The study population included respondents who had answered all relevant questions (Fig. 1).

Basic descriptive analysis was used to study the involvement of each personal and professional relation when experiencing each of the following six symptoms: frequent urination, nocturia, stress incontinence, urge incontinence, incontinence without urge/stress and difficulty emptying the bladder. Subsequently, the three incontinence symptoms were merged into one symptom category named ‘incontinence’, giving a total of four LUTS (frequent urination, nocturia, incontinence and difficulty emptying the bladder). The merge was due to few observations in each symptom group, and the category included the incontinence symptoms if any of the three incontinence symptoms were reported as bothersome. The study sample was stratified into four age groups: < 40 years, 40–59 years, 60–79 years and ≥ 80 years.

Regarding the available social network, individuals were categorized as having no available social network if all the following responses were chosen: never/less than once a month being in contact with others, having no available individuals who can help, often being alone when desiring to be with others and never/almost never having a person to talk to in case of problems.

We used multivariate logistic regression models to analyse possible associations between involvement of personal and professional relations and specific characteristics (age, number of symptoms, available social network and personal relations). Crude and adjusted odd ratios were calculated. Adjustments were made for possible confounders: age, number of symptoms and available social network. For the logistic regression analyses we categorized the personal and professional relations into five groups: [1] the GP, [2] another doctor, [3] other healthcare professionals, [4] family members and [5] personal relations other than family. Other healthcare professionals comprised all professional relations selectable apart from the GP and another doctor, family members included spouse/partner, parents and children and personal relations other than family encompassed colleague/classmate, friend, neighbour and others.

This population-based study comprised 22,297 randomly selected men from the general population and of those 4885 reported at least one bothersome LUTS yielding a prevalence of 21.9%. Among men experiencing bothersome LUTS, 23.5% involved neither professional nor personal relations regarding their symptom(s). Involvement of personal relations was most common and was associated with a two to four-fold increased odds of involving the GP, another doctor and other healthcare professionals. The most frequently involved professional relation was the GP followed by another doctor and the odds of involving either of these were highest among the oldest group of men. Men categorised as having no available social network had higher odds of involving the GP and other healthcare professionals regarding frequent urination. For the remaining LUTS a similar tendency was observed, however not statistically significant.

This study was a large nationwide cross-sectional study including 22,297 male respondents. To our knowledge such a large population-based study has not previously been conducted concerning personal/professional relations and network activation among men with bothersome LUTS. However, the cross-sectional nature of the study is a limitation. Longitudinal data would allow the temporal sequence of variables to be established more strongly.The response rate of 49.8% was consistent with previous studies [3, 20]. However, it is unknown whether individuals experience symptoms are less or more inclined to participate in the study and thus an underestimation or overestimation of the prevalence cannot be eliminated.

The web-based questionnaire was not available in a paper version, which may have prevented some invitees from participating in the study, especially the elderly. The possible selection bias was minimized by offering the possibility of conducting the survey as a telephone interview to people with no internet access. The random selection through CRS was also used to eliminate the risk of selection bias.

Information on the symptom experiences and who the respondents involved regarding their symptoms were self-reported and since LUTS might be associated with shame and embarrassment [11] it is possible that an underestimation of the prevalence of LUTS is present. However, the underestimation is presumably minimized by the web-based design because of an increased perception of anonymity. The invitees were asked to recall symptom experiences within the preceding 4 weeks, and whether they had talked to a personal or professional relation regarding these symptoms at any time. The 4 weeks recall period was used to obtain statistically precise estimates while still assuming the participants could recall symptoms accurately. However, recall bias cannot completely be eliminated in questionnaire studies. Some may have forgotten to report a symptom or a relation they had involved because the symptom experience turned out to be inconsequential or simply due to memory decay. Others might have misplaced symptom experiences outside the timeframe of the study due to severity of the symptoms or because they had already involved a personal/professional relation about them. As only individuals bothered by LUTS are included we believe that this risk of recall bias is negligible [21].To obtain further information about the symptom disclosure more response options could have been presented. For instance, the use of social media or the internet might have been a frequent reason not to contact the GP among the young respondents.

This paper investigates bothersome LUTS which were defined as being of moderate to extreme concern or influence. This definition was chosen because LUTS often are of benign origin and therefore it might be unnecessary for men with no bother from their symptoms to seek medical attention. The term bothersome was a construction made in the author group based on the literature and clinical experience and is described more in detail elsewhere [19]. This construction, however, still needs further validation in the population.

To be defined as ‘bothered’ respondents could be either moderate to extremely influenced by the symptoms without being concerned or vice versa. It may have influenced the results that respondents categorised as bothered by LUTS could be moderate to extremely influenced by the symptoms without being concerned or vice versa. A diagram of the distribution of bothersome is shown in Rubach et al. [19] and it reveals that more than half of the symptoms were of both moderate/extreme influence and moderate/extreme concern. It is plausible that men being both extremely concerned and extremely influenced by LUTS are more likely to involve professional or personal relations compared to men being moderately influenced in their activities but with no or little concern. Whether influence or concern is the greater driver for involvement of others is yet to be determined.

We hypothesized that involving a personal relation may influence the decision to seek healthcare with bothersome LUTS. However, the questionnaire did not give any information about the chronological order in which the relations were involved. Therefore, we can only describe whether a relation was activated or not. Furthermore, the study did not reveal the quality or content of the contacts made and thus it is unknown whether a potential advice from the personal network is in favour of help-seeking or not. If the advice from the personal network is inadequate it may prevent the patient from contacting the GP and thus delay medical evaluation.

We intended to analyse the impact of having an available, functional network on the help-seeking process, and therefore presumed that individuals having “contact less than once a month” and “almost never having a person to talk to” had a very sparse network. In the dichotomized analyses they were therefore categorized as having no available network.

Around one-third of LUTS were discussed with the GP in this study, which is lower compared to a study by Pescosolido, who found that the GP was involved in 85.4% of the “illness episodes” registered in a retrospective survey [13]. The high involvement of the GP in the study by Pescosolido was expected due to the way “illness episodes” were selected. We found that a minority of symptoms are discussed with the GP, which is in concordance with previous studies [10, 20]. Most LUTS are of benign origin and therefore treatment should primarily be targeted men with bothersome symptoms. As our survey was conducted in a gatekeeper system with free access to the GP, a high utilization of the GP was expected among our study population. However, more than half of men experiencing bothersome LUTS had not involved any professional relation. In comparison, a population-based study by Boyle et al. [22] found that 40.9–77.5% of men with bothersome incontinence involved a doctor regarding their symptom, which is higher compared to our results (34.7%). Boyle et al. [22] had a different study population of men aged 40–79 years compared to this study, where men above 20 years were included. In Boyle et al. [22] bothersome is defined based on the impact on daily life by using the BPH Impact Index whereas this present study included symptoms being either concerning and/or influencing everyday life. Further, no timeframe for experiencing the symptoms was defined in the study by Boyle et al. Norby et al. [3] found that 9.2% of men > 50 years of age with voiding problems had involved a doctor within the past 2 years, which is lower than our findings for difficulty emptying the bladder (38.4%) and may be caused by the longer timeframe, the different study populations and the severity classification. Further, they found that elderly men involved the doctor more often than the younger men [3]. This matches our study where a higher proportion of men above 80 years had involved the GP (44%) compared to men < 40 years of age (13.7%). The more frequent involvement of the GP among elderly men could possibly be caused by the fact that elderly men consult their GP more often for other reasons which provides more opportunities to present the LUTS. Likewise, the more frequent consultations could provide the elderly with a better relation to their GP and thus it could eliminate the shame and embarrassment associated with LUTS. In addition, younger men might be more likely to search the internet for advice instead of consulting the GP compared to the elderly.

Surprisingly, 77.9% of men < 40 years of age with bothersome LUTS did not involve any professional relations. Various barriers to healthcare seeking among men in general have been suggested by Yousaf et al. [23], and might be part of the explanation for this finding. Yousaf et al. found that there were several psychological and contextual factors discouraging men to seek medical help. Among these were embarrassment, anxiety and fear along with lack of knowledge about symptoms, treatments and services and viewing symptoms as minor and insignificant [23]. Furthermore, poor communication and lack of time was mentioned as important factors for initiating medical contact [23]. To what extent these factors affected our results is uncertain, but nevertheless a possible explanation for the finding.

In the present study the spouse/partner was the most frequent relation involved in general and was involved in more than half of all the cases. Friends and children followed the spouse/partner as the most common personal relation involved. These findings are supported by Roe et al., who found that spouse/partner was most frequently involved, followed by family and friends, for patients with urinary incontinence [24]. However, a lower prevalence of involving personal relations was found by Roe et al. as 47–49% had involved the spouse/partner compared to 58.4–67.6% in our study. The involvement of a friend was 25–32% in Roe et al. compared to 8.9–11.3% in our study [24]. Roe et al. included both men and women and the study were conducted in the UK, which could be the explanation for the difference in the results.

We found that odds of involving the GP, another doctor and other healthcare professionals were two to four-fold higher when a personal relation was involved. The findings are comparable to the study by Eriksson et al. [12], who found that being advised by others to seek medical attention increased the likelihood of seeking primary healthcare approximately five-fold. This might indicate that personal relations can act as a trigger of healthcare seeking.

In the present study, men without an available social network more often involved the GP about frequent urination compared to men with an available social network. Similarly, a paper with results from three studies showed that low social group connectedness was associated with a higher frequency of primary care attendance [15].

In the present study, men without a social network were more likely to involve a personal relation other than family compared to men with an available social network. The explanation for this could be that men without an available social network disclose their symptoms to a neighbour or colleague whom they cannot rely on in case of illness. Furthermore, it could indicate that family was interpreted as the available social network for most men.