Prevalence of constipation and use of laxatives, and association with risk factors among older patients during hospitalization: a cross sectional study

Constipation is a common gastrointestinal condition, that most people have experienced at least once in their lifetime. Even though constipation is a common health condition, the existence of various definitions may partly explain why constipation is often under-diagnosed or undertreated [1].

The prevalence of constipation among older adults increases with age [2]. Constipation can lead to complications such as fecal impaction or incontinence, hemorrhoids, volvulus, or rectal prolapse [3].

Many older people experience constipation as a bothersome and distressing health condition with negative impacts on their quality of life and social activities [4]. Healthcare professionals experience that the assessment of bowel health is not prioritized, when older patients are admitted to hospital care, since the focus is on the primary symptoms related to the cause of admission [5]. A hospital stay can be challenging for older patients due to multimorbidity, in combination with changes in their environment, such as inactivity, changed eating- and drinking habits, lack of privacy and being dependent on help. This can lead to altered bowel habits and an increased risk of becoming constipated [6]. Previous research has found that constipation and laxative use are associated with higher mortality [7], and that multimorbidity might increase the likelihood of becoming constipated [8]. This can, in turn, lead to increased health care costs [9]. Healthcare professionals have also stated that they often underestimate the severity of constipation as perceived by the patient [10] and patients and physicians might have different perception of constipation [11].

The most commonly used definition of constipation is based on the Rome criteria (Mearin et al., 2016). This definition comprises a set of clinical symptoms recognized as the gold standard for constipation and is often used in clinical research [12] (Table 1). The condition constipation is described as “another functional intestinal disorder” in the International Statistical Classification of Diseases and related Health Problems (ICD-10) [13]. ICD-10 is a diagnostic classification standard used globally in clinical practice to systematically describe and communicate medical diagnoses and health-related problems. From a nursing point of view, constipation is also addressed as standardized nursing diagnoses, systematic descriptions of health conditions, e.g. the North American Nursing Diagnosis Association (NANDA) classification system. Nursing diagnosis of constipation, as an actual or a potential health problem, is defined with identified related factors or risk factors and also provides characteristics of the condition, e.g. abdominal pain, or risk factors, such as change in eating habits [14]. Further, there are several assessment scales that have been developed to support the identification of constipation, such as the Constipation Assessment Scale [15] or the Bowel Function Index [16]. In some studies the prescription of laxatives has been used as an indicator of constipation [17].

In a systematic review including studies worldwide, the prevalence of constipation varied between 1 and 80% among adults in different settings and using different definitions [18]. In a Swedish study, an increase in prevalence of constipation in nursing home settings was reported, from 36% in 2007 to 40% in 2013, based on healthcare professionals responses to a three item questionnaire [19]. With a definition of “no bowel movements in three days or hard stools” [20] a study found the prevalence of constipation to be 23% among nursing home residents, while 67% were prescribed laxatives for regular use. Few studies have examined the prevalence among older patients admitted to hospitals. However, one Danish study used the Constipation Assessment Scale and reported that 39% of patients indicated symptoms of constipation on admission and 43% had developed symptoms of constipation during the first three days of hospitalization [21].

Increased age is considered to be one of the most important risk factors for developing constipation, and among older people, constipation is often caused by several health conditions that sometimes interact [22]. Associated factors for constipation include physical and mobility impairments, poor appetite and impaired nutritional status [23]. Polypharmacy and specific medications, such as opioids and anticholinergics, are also associated with increased risk of constipation [19, 24]. There are also gender differences, with females being more affected than males [22]. Constipation among nursing home residents is associated with a other health risks such as risk of falls or risk of developing pressure ulcers [25].

The aim of this study was to describe the prevalence of constipation, the use of laxatives, and the association between constipation and demographic factors, risk assessments and prescribed medications, among older hospitalized patients.

The results in this study, showed that the estimated prevalence of constipation among older patients admitted to hospital differs depending on the method used for identifying the condition. There are currently no universally accepted criteria for the diagnosis of constipation in the hospital environment [31]. Based on the ICD-10 diagnosis, 6% of the patients had been diagnosed with constipation. The prevalence markedly increased to 65% when notes including signs and symptoms of constipation were analyzed. Previous research has also found levels of disagreement between the ICD-10 diagnosis and documentation of signs and symptoms related to other health conditions in EHRs [32]. It has been shown that when nursing diagnosis and medical diagnosis are combined, the results better predict risk of complications, length of stay and more [33, 34]. This might also be the case in the context of constipation. Another explanation might be that the ICD-10 diagnosis documented in the patient file where only those related to the current admission. The use of the ICD-10 diagnosis for constipation resulted in low prevalence, whereas studies measuring constipation where it is self-defined by patients tend to report much higher prevalence rates [35]. In this study, 60% of patients were prescribed laxatives, indicating more than ten times the prevalence of constipation than according to the ICD-10 diagnosis. In order to evaluate this difference, we must however have in mind that some patients received laxatives to prevent constipation, while other received laxatives to treat constipation. Our results also showed that 32 (10%) patients were prescribed laxatives, without any signs and symptoms of constipation or ICD-10 diagnosis being documented in the EHR (Fig. 1). This difference in prevalence may indicate that constipation is insufficiently diagnosed and documented in the older patient population.

The prevalence of prescribed laxatives increased from 29% at admission to 60% during hospitalization, which is also in line with previous studies and could be related to changes in the patient’s situation, such as diet, medication, immobility, loss of privacy and unfamiliar surroundings [1, 17]. Pharmacological treatment of constipation with prescribed laxatives was most common, and the use of non-pharmacological treatments was rare. A previous study has also reported that non-pharmacological treatments were rarely used [1] and the general explanation suggests that a laxative is chosen since a prompt solution to the problem might be needed [36]. Non-pharmacological treatment options could be appropriate for nurses to use in clinical practice as could practical complementary methods for nurses to use in clinical practice, however; the evidence is still inadequate regarding the effect of these interventions [37]. A systematic review evaluating the effect of non-pharmacological interventions among older people living in nursing homes concluded that making comparisons between study findings is challenging due to the differences in frequency, duration and content of the interventions [38].

Documented signs and symptoms of constipation were significantly associated in the bivariate analysis with risk of developing pressure ulcers, longer length of stay, being prescribed laxatives and opioids. Being prescribed opioids and longer length of stay remained a significant risk factor in the logistic regression model. In previous studies among older persons, risk of developing pressure ulcers and risk of falling have been identified as being associated with constipation. However, in our study this was not the case. One explanation for this difference could be related to severe mobility impairment and functional impairment among the nursing home residents under investigation [25]. Furthermore, when using prescribed laxatives as a dependent variable, age, prescribed opioids and length of stay were identified as associated factors in the regression analysis. No significant associations with constipation were detected for underlying health conditions in our study, which is in line with previous research [25]. This finding indicates that comorbid health conditions may have a lower impact on the development of constipation, when measured by the CCI. However, in contrast, a previous review reported that underlying medical conditions were one of the most common risk factors for constipation [39]. It would therefor be of importance, to prevent constipation and identify persons in risk, to explore further which specific medical conditions poses a risk for the older person.

In our study, 50% of the patients were prescribed opioids and nearly one third (31%) of those were not prescribed any laxatives. Polypharmacy is rather common among older people and many types of drugs, e.g. opioids, anticholinergics, diuretics, proton pump inhibitors, iron and calcium supplements, are associated with an increased risk of constipation especially, among older people who are often prescribed opioids [40, 41]. It is recommended that laxatives should be prescribed preventatively in patients receiving opioids, but one study reported that physicians relied on other clinicians outside their purview, such as the primary care provider, to take care of problems with constipation or waited until the patient expressed concerns during follow-up assessments [42]. Although attempts have been made to increase awareness about co-prescription of laxatives in conjunction with opioids, this has shown only a small effect in terms of outcome [43]. The NANDA international terminology system aims to ensure that the nursing process is systematically and unambiguously documented, but it is rarely used as a basis for documentation in healthcare [44]. The documentation is often inadequate, with only isolated and fragmented elements included in the EHR. This was also found in our study where outcomes were documented mostly related to information, such as the evacuation of feces, without any description of the process in connection with defecating. The Venn-diagram also demonstrates a contradictory picture of identification of constipation using ICD-diagnosis, documented signs and symptoms, and prescription of laxatives. Constipation was documented according to all three data sources in the EHRs for 5% of the patients. The majority (n = 143; 45%) of the patients with documented signs and symptoms of constipation also had a documented prescription for laxatives, but they did not have an ICD-diagnosis. One patient with an ICD-diagnosis was prescribed laxatives, but without any documented signs or symptoms of constipation. Forty (12%) patients with documentation of signs and symptoms had neither an ICD-diagnosis nor prescribed laxatives documented. Problems with documentation in EHRs can be related to structural deficiencies, poor compatibility between different systems, and the use of different vocabulary to describe the nursing process [44]. The use of a variety of terminologies can lead to misinterpretation of information between healthcare professionals and can therefore potentially jeopardize patient safety [44]. It is of great importance that the documentation is performed in a structured and standardized manner, since this can facilitate better communication between healthcare professionals and enhance the quality of care for patients. It has been previously described that healthcare professionals reported not having access to standardized valid assessment scales for identifying constipation and that this was perceived as a limitation in their clinical work. This contributes to assessments taking place in an unstructured way based on healthcare professionals’ individual experience [5].

The estimated prevalence of constipation varied to a large extent depending on the data source used to identify the condition. It is noteworthy, that bowel health, which is an important subject for people’s wellbeing in daily life does not seem to be prioritized in health care, despite the existence of validated assessment scales, and guidelines regarding the management of constipation. Apparently, more needs to be done to support healthcare professionals with existing valid and reliable methods for the identification and management of constipation among older patients.