Introduction

It has long been observed that people with schizophrenia have higher all-cause morbidity and mortality rates when compared to the general population [1]–[3], and the gap has been widening in recent decades [4]. Two previous studies found that over half of their subjects with schizophrenia had been diagnosed with at least one medical co-morbidity [5], [6]. In another previous study with a dataset including subjects with schizophrenia or schizoaffective disorder and controls who filed at least one claim for medical services, it showed that the mean age of the cases was 40.2 years and one-third of the cases had three medical co-morbidities or more than controls [3]. Frequently reported medical co-morbidities include infections, obstetric complications, cardiovascular diseases, respiratory, endocrinologic, and metabolic illnesses, and alcohol and other drugs concurrent disorders [7], [8].

Although characterized by higher overall morbidity and mortality rates, people with schizophrenia have been demonstrated to have negative associations with various medical conditions. For example, after reviewing 14 epidemiological studies, Eaton et al. concluded that, despite their various methodological limitations, their studies conclusively proved that there was a lower than expected prevalence of rheumatoid arthritis among people with schizophrenia [9]. Additionally, Barak et al. analyzed a cohort of 3,226 people with schizophrenia, and observed a standardized incidence ratio (SIR) of 58% (p<0.05) for all cancers in the cohort was observed, compared to the rates in the general population [10].

Although a possible association exists between schizophrenia, polydipsia [11], [12] and urinary retention [13], little attention has been given to urinary diseases further downstream. Urinary systems might be more vulnerable among people with schizophrenia [14]–[16], urinary calculi, as one of the most frequently reported diseases of the urinary system and its detrimental consequences, requires more investigation.

A lifetime risk of urolithiasis exceeded 12% in males and 6% in females [17], [18]. Rising trends have been suggested for both the prevalence and incidence of urinary stones [17]. Furthermore, if left untreated, about 30–40% of patients develop another stone within five years of the initial episode [19], [20]. Whereas, with proper intervention, recurrence rates have been reported to be reduced up to 85% in randomized trials [21], [22]. This indicates that there is a need for the prompt detection of urinary calculi to ameliorate symptoms, decrease reoccurrence rates, and reduce healthcare costs. Assessment of the risk of urolithiasis is critical in the case of vulnerable populations, such as schizophrenia, in order to draw appropriate clinical attention and provide proper health care, if there is any relationship identified. Nevertheless, no direct evidence to date has been reported on the link between schizophrenia and urinary calculi.

In pursuit of these goals, this study utilized a population-based case-control study design to investigate the possibility of an association between schizophrenia and the occurrence of urinary calculi. The prevalence and risk of schizophrenia were examined among patients with and without urinary calculi in a claims dataset in Taiwan. Further analysis included stratification by age and sex.

Methods

Results

Of the 323,790 study sample, the mean age was 47.8 (±15.6 years); about 56% were aged under 50 years (Table 1). There were significant differences between patients with and without urinary calculi in terms of the monthly income, level of urbanization or the geographic location of the patient's residence, hyperlipidemia, diabetes, hypertension, hypothyroidism, hyperthyroidism, obesity, and polydipsia (p<0.001). But there is no significant difference in the distribution of alcohol abuse/alcohol dependence syndrome between cases and controls (p = 0.297). Patients with urinary calculi were more likely to have monthly incomes >NT$15,840 and to reside in the eastern part of Taiwan than patients without urinary calculi.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic characteristics of patients with urinary calculi and controls in Taiwan in the year 2002 (n = 323,790).

https://doi.org/10.1371/journal.pone.0056942.t001

The prevalence of prior schizophrenia in patients and controls is shown in Table 2. A total of 3,119 (1.0%) subjects had been diagnosed with schizophrenia prior to the index date. This included 0.7% of the patients with urinary calculi, and 1.0% of the controls. The conditional logistic regression analysis, conditioned on patient sex and age group, indicated that the odds of having previously been diagnosed with schizophrenia for patients with urinary calculi were 0.69 (95% CI = 0.62–0.76, p<0.001) when compared with controls after adjusting for patient's monthly income, and level of urbanization, and the geographic location of the patient's residence, hyperlipidemia, diabetes, hypertension, hypothyroidism, hyperthyroidism, obesity, alcohol abuse/alcohol dependence syndrome, and polydipsia.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Crude and adjusted odds ratios for urinary calculi stratified by the presence/absence of schizophrenia among the sampled patients.

https://doi.org/10.1371/journal.pone.0056942.t002

Table 3 presents the association of urinary calculi with schizophrenia stratified by sex. Among the male and female cases, patients with urinary calculi were 38% (95% CI = 0.55–0.71, p<0.001) and 20% (95% CI = 0.69–0.94, p<0.01), respectively, less likely to have a prior schizophrenia diagnosis as patients without urinary calculi.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Odds ratios for schizophrenia among patients with urinary calculi and controls, by sex.

https://doi.org/10.1371/journal.pone.0056942.t003

Table 4 presents the ORs for prior schizophrenia stratified by age group. Apart from the <30 group, when compared with controls, patients with urinary calculi were consistently less likely to have prior schizophrenia across all the age groups. The adjusted OR for prior schizophrenia for the <30 age group was 1.91 (95% CI = 1.45–2.53, p<0.001) times that of patients with urinary calculi than of controls.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Odds ratios for urinary calculi among patients with schizophrenia and comparison group, by age group.

https://doi.org/10.1371/journal.pone.0056942.t004

Discussion

To the best of our knowledge, this is the first study to identify an association between schizophrenia and urinary calculi. Our results showed that a prior diagnosis of schizophrenia was associated with a 30% decrease in the occurrence of urinary calculi, after adjusting for monthly income, urbanization level, and the geographic location of the patient's residence. Males (36%) exhibited an even further reduced risk than females (17%). The most pronounced difference was found among older age groups (a reduction of 47% in those aged >69, 42% in those aged 50–69, and 35% in those aged 30–49). Conversely, a previous diagnosis of schizophrenia increased the risk of urinary calculi by 92% for patients aged less than 30 years after adjusting for potential confounding effects.

A number of reports have suggested that people with schizophrenia have higher than expected rates of some physical co-morbidities [1]–[3], while lower than expected rates of other physical illnesses [9], [10]. An array of frequently reported medical co-morbidities in schizophrenia included cardiovascular, neurological, genitourinary, respiratory, and gastrointestinal diseases [7], [23]. Whereas, people with schizophrenia were also found to possess a reduced risk for developing either rheumatoid arthritis or lung cancer [24]. Our study suggested that urinary calculi as another disease that might be negatively associated with schizophrenia.

It is worth noting that the risk reduction in urinary calculi was more remarkable among males than females. Sex hormones may be involved in the gender differences [25], though future studies are needed to further elucidate the reasons.

The reduced risks were more apparent among elder patients than younger ones. While the rationale for the trend of decreased risk is as yet unknown, some possibilities might be suggested. Renal dysfunction experienced in those with undiagnosed urinary calculi due to schizophrenia (via possible decrease in pain level or help seeking tendency) could be responsible for a disproportional increase in mortality. Consequently, a disproportionately small number of those with schizophrenia and urinary calculi could be observed in the older age groups. Furthermore, although most psychotropic medications are not metabolized or excreted by the kidneys, a large number of medications taken by those with schizophrenia could potentially exacerbate renal dysfunction. Also given that diabetes and hypertension are highly comorbid with schizophrenia [7], their combination with the renal dysfunction caused by urinary calculi could be particularly detrimental. It is also possible that people at a younger age were less affected by potential confounders (e.g., living conditions, dietary habits, history of previous medication). More studies should be devoted to explore the role that age plays in the link between schizophrenia and urinary calculi.

The mechanism of the negative relationship between schizophrenia and urinary calculi is unknown; yet preliminary evidence suggests several potential explanations. First, thyroid dysfunction, particularly hypothyroidism, may be contributory. Existing literature suggests that primary hyperparathyroidism or hyperthyroidism were conditions that may increase the risk of calcium stone formation [26]–[28]. However, hypothyroidism was found to be nearly three times as prevalent among people with schizophrenia than among non- schizophrenia people [3], [7]. Hypothyroidism might cause the risk of stone formation to subsequently decrease and deserved further investigation.

In addition, although speculative, as coffee, tea, beer and wine are associated with a decreased risk of stone formation [29], higher than average rates of alcohol use and caffeine consumption among people with schizophrenia might contribute to their reduced risks in urinary calculi [23].

Finally, many co-morbid illnesses may be undiagnosed or misdiagnosed in people with schizophrenia [7], [23]. Barriers to care delivery, such as the patients' poor ability to recognize medical conditions and disinclination to seek medical help, might cause healthcare professionals to overlook co-existing medical problems. All the above factors may contribute to the reduced risk of being diagnosed with urinary calculi among these vulnerable individuals.

An important strength of our study is that it utilized a large case-control population-based design, leaving little room for selection and non-response biases. By utilizing a claims dataset, this study largely avoided recall bias, a major concern in case-control studies in general.

However, the findings of this study need to be interpreted within the context of three limitations. First, as stone disease is rarely fatal and is characterized by variable pain intensity, calculi may often prove asymptomatic in the absence of obstruction. The diagnosis of urinary calculi in the claims dataset of our study was thus subject to a patient's individual pain tolerance and healthcare utilization behavior. Second, the social stigma associated with schizophrenia and the lack of insight into schizophrenia from seeking medical attention might prohibit patients with psychosis from receiving an appropriate diagnosis or treatment, which would contribute to a possible misclassification bias. Nevertheless, schizophrenia is classified as a major catastrophic illness under the National Health Insurance scheme. In Taiwan, once the diagnosis of a major catastrophic illness has been verified, the co-payment for all subsequent outpatient and inpatient healthcare utilizations is waived. It is thus in the patients' best interest that schizophrenia is faithfully diagnosed and appropriately treated. To further increase the diagnostic validity of schizophrenia in this study, we also confirmed that a prior diagnosis of schizophrenia could only be included if it was received in an inpatient setting, or appeared in at least two ambulatory care claims coded before the index date. Third, some important variables such as a family history of stone disease [30] and the patient's body mass index [31] and diabetes status [32], which are likely risk factors of urinary calculi, were unavailable in our LHID2000 claims dataset. Therefore, our findings could be compromised. Finally, the study population mainly consisted of Taiwanese Chinese, and therefore, the results may lack generalizability to other ethnic populations.

In sum, this is a pioneering study in exploring the association between schizophrenia and urinary calculi. Replication of our findings is warranted in other regions or countries. More studies will be needed to examine whether the apparent reduction in the risk of urinary calculi among schizophrenia is fictitious, and merely an artifact of under-diagnosis, or if it reflects a true decreased risk of urolithiasis among people with schizophrenia.Thus, if under-diagnosis is an issue, dissemination of our negative findings to healthcare professionals may raise awareness and help compensate for a possible deficit of medical attention paid on the urinary health of patients with schizophrenia. Prolonged hospitalizations, treatment failure, and decreased quality of life due to under-diagnosed may thus be avoided. Future studies are needed to elucidate the mechanisms by which schizophrenia negatively associates with urinary calculi.