Incidence of “new-onset” constipation and associated factors during lockdown due to the coronavirus-19 pandemic

Background: A minimum of physical activity and low liquid intake are factors that have been associated with constipation. The health emergency brought on by the COVID-19 pandemic has resulted in adopting behavior, such as sheltering-in-place (less mobility) and dietary changes, creating a scenario we believe to be an adequate model for examining the appearance of symptoms of constipation and its associated factors. Methods: : A cross-sectional and descriptive study was conducted on an open population, applying an electronic survey (4 weeks after lockdown due to COVID-19 in Mexico) to evaluate: demographic characteristics, physical activity, water and fiber intake, appearance of constipation symptoms (including stool consistency), and quality of life. Key results: Out of 678 subjects evaluated, 170 (25%, 95% CI 21.7-28.4) developed symptoms of “new-onset” constipation, with a significant decrease in the number of daily bowel movements (p<0.05) and hardness of stool consistency (p<0.05) during lockdown. The multivariate analysis (logistic regression) showed that female sex (p=0.001), water intake (p=0.039), and physical activity (p=0.012) were associated with “new-onset” constipation. Conclusions: : Reduced physical activity and less water intake due to mobility restrictions imposed for epidemiologic reasons, such as a pandemic, can lead to symptoms of “new-onset” constipation.


Introduction
Constipation is one of the most frequent gastrointestinal disorders and is estimated to affect up to 14% of the general population. [1,2,3,4] The pathophysiology of constipation is complex and multifactorial. [1,5,6] Among the factors that contribute to the appearance or worsening of constipation symptoms are those that are considered unmodi able, such as female sex, advanced age, medication use, and the association with different comorbidities, such as neurologic diseases. [7] On the other hand, there are numerous factors, associated with lifestyle, that are considered modi able, such as low ber intake, [8,9] inadequate water intake, [10,11,12] sedentary lifestyle, and reduced physical activity. [12,13] Nevertheless, evidence on the theme is limited and controversial, given the di culty in conducting prospective studies on the topic. Some studies have demonstrated that physical activity has no effect on the motor activity of the colon, whereas others report the opposite, showing it can depend on the frequency, intensity, and duration of exercise (for example, marathon running). [11,13,14] Symptoms of constipation induced by immobility in geriatric patients have been shown to decrease with the regular performance of physical activity. In a unique study on the subject, Iovino et al. [15] demonstrated that prolonged physical inactivity (for 35 days, in 10 healthy volunteers) triggered constipation in 60% of the study individuals.
The COVID-19 pandemic, caused by the SARS-CoV-2 coronavirus and declared in March 2019 by the World Health Organization (WHO), has resulted in an unprecedented public health crisis worldwide. [16] This health emergency has produced substantial changes in social, economic, and cultural spheres. In Mexico, contingency measures (reduced essential activity, physical distancing, stay-at-home orders, etc.) were established on April 21, 2020. Evidence suggests that the obligatory social self-isolation impacts the lifestyle of the population, forcing the adoption of unhealthy dietary habits (increased ultra-processed and re ned foods and decreased fruit and vegetable intake), sedentary behavior, and sleep disturbances. [17] At present, there are no reports in the literature that establish a relation between a change in bowel movement habit and being under the current lockdown. Thus, the aim of our study was to evaluate the incidence of symptoms of constipation and associated factors during the lockdown implemented to contain the spread of COVID-19 in Mexico.

Study design and population
A descriptive, cross-sectional study was conducted on an adult Mexican population (above 18 years of age), in which volunteers were asked to take a survey that was distributed on social media throughout Mexico. The survey was designed and applied utilizing the "Google Forms" tool. Because the lockdown (phase 3 of the pandemic) was put in place in Mexico on April 21, 2020, the survey became available on May 18, 2020 (four weeks later), after obtaining informed consent from the volunteers that wished to participate, along with their authorization of personal data use, ensuring data privacy.
There was no incentive for completing the survey.

Data collection
The following data were collected in the survey applied to each of the participants: a. Sociodemographic characteristics: Sex, age, weight, height, body mass index (BMI), educational level, comorbidities, and medication b. Physical activity before and after the lockdown: Based on the international physical activity questionnaire-short form (IPAQ-SF), [18]  c. Water and ber intake before and after the lockdown: Water intake was evaluated according to the 4 following categories: 1) < 500 ml per day, 2) 500 ml to 1 liter per day, 3) 1 to 2.5 liters per day, and 4) > 2.5 liters per day (n, %). Regarding ber, the participant was asked if intake was on a regular basis and in what form it was consumed (cereals, fruits and vegetables, supplements in capsules, supplements in powder).
d. Constipation symptoms, stool consistency, and bowel movement frequency: The participants were asked in binary form (yes or no) if they considered themselves to have constipation before the lockdown. The participants with prior symptoms of constipation were excluded. Those that answered that they had not previously presented with constipation were asked, "Do you think you have developed constipation symptoms since the lockdown was put in place?" In addition, based on the questions for de ning chronic constipation from the Rome IV modular questionnaire, the following symptoms were evaluated: straining, pushing force needed to defecate, stool consistency types 1 and 2, according to the Bristol stool form scale, in more than 25% of the bowel movements, and use of digital maneuvers. The participants were also asked the number of bowel movements they had per day and per week, and the average stool consistency, according to the Bristol scale, in the past e. Quality of life in relation to constipation. The patient assessment of constipation-quality of life (PAC-QOL) questionnaire was used because it is the most widely validated, speci c tool for measuring quality of life in patients with constipation. [19] It has been validated and used in Spanish [20] and consists of 28 items grouped into the following subscales: (a) physical discomfort, b) psychosocial discomfort, c) preoccupation, and d) treatment satisfaction, and includes an overall score. Scoring is on a Likert scale of 5 points, from 0 (nothing/never) to 4 (extremely, always), and a lower score re ects a better quality of life.

Statistical analysis
Incidence (new cases) of constipation after the lockdown ("new-onset" constipation) was calculated and the 95% CI was reported. A comparative analysis of the categorical variables and continuous variables was carried out between the participants with "new-onset" constipation and those with no constipation during the lockdown, utilizing the Student's t test, the Mann-Whitney U test, the chi-square test, or the Wilcoxon test, as appropriate. Statistical signi cance was set a p < 0.05. A multivariate analysis was then performed utilizing logistic regression, employing the enter model. The analysis of the results was carried out through descriptive statistics with the IBM® SPSS Statistics® version 22 program. The protocol followed the Declaration of Helsinki stipulations for biomedical research and was approved by the institutional committee (IIMB-2020-008)

Results
General population characteristics A total of 772 questionnaires were received, 94 of which were excluded (when more than one response was selected, survey takers were from countries other than Mexico, surveys were incompletely lled out, etc.), leaving 678 study participants. Of those subjects, 405 were women (59.7%), 273 (40.3%) were men, and the mean participant age was 31.21 ± 11. 9 years (range 18-74).
In those subjects, the median of bowel movements per day decreased from 2 (range 1-6) to 1 (range 1-6) after the lockdown (p=0.00001). Likewise, the median of bowel movements per week decreased from 9 (range 3-35) to 7 (range 1-28) after the lockdown ( Figure 2 Factors associated with "new-onset" constipation before the lockdown Table 1 shows the baseline characteristics and associated factors (before the lockdown) in the 170 subjects with "new-onset" constipation and the 508 subjects that had no constipation symptoms. A signi cantly higher percentage of women than men developed "new-onset" constipation (p=0.0001). The presence of comorbidities, water and ber intake, physical activity, number of bowel movements, and Bristol scale-assessed stool consistency were similar between the two groups before the lockdown.  Factors associated with "new-onset" constipation during the lockdown According to the univariate analysis, the patients that developed "new-onset" constipation had less physical activity during the lockdown, and if performed, it was at a lower intensity. A higher number of subjects with "new-onset" constipation drank less than 1 liter of water per day (30% vs 20%, p=0.000). No differences were found in the duration or number of days of physical activity or in ber intake during the lockdown between the two groups ( Table 2). In the multivariate analysis (logistic regression), female sex, water intake, and physical activity were associated were associated with the development of "new-onset" constipation (Table 3). Quality of life according to the PAC-QOL questionnaire Figure 3 shows the means of the 4 subscales and the overall score of the PAC-QOL questionnaire ( Figure  3). The mean score for the physical discomfort subscale was 2.16 ± 0.72, for the psychosocial subscale was 1.74 ± 0.6, for the preoccupation subscale was 2.98 ± 0.8, for the satisfaction subscale was 3.46 ± 0.9, and for the overall score was 2.3 ± 0.6.

Discussion
Our study ndings are novel, in the sense that for the rst time, 1 out of every 4 (25%) subjects whose physical activity was forcibly reduced as a consequence of the COVID-19 pandemic were shown to have developed "new-onset" constipation. The lockdown and mobility restriction measures put in place to contain the spread of COVID-19 created a "natural" model, in which, according to our results, reduced physical activity and lower water intake were determining factors for the appearance of constipation.
In the subjects with "new-onset" constipation, there was a signi cant decrease in the number of bowel movements, as well as signi cantly harder stools during the lockdown. Very importantly, the fact that we found an increased occurrence of harder stools, utilizing the Bristol scale, which is an objective surrogate measure of colon transit, indicates that colon transit time was slower in those subjects during the lockdown.
The role of mobility and physical activity in relation to chronic constipation is a subject of debate, [12,13] but our study showed that reduced physical activity was a determining factor for the appearance of constipation symptoms. Numerous studies have reported that prolonged immobilization is a trigger for the appearance of symptoms of constipation. [21,22] Even though those ndings have been described in chronically ill patients, little is known about reduced mobility and the appearance of constipation in a population that previously was apparently healthy, such as the one we evaluated. Our results are similar to those reported by Iovino et al., [15] in a "new-onset" constipation model. Those authors demonstrated that strict bedrest for a period of 35 days, adhered to even to perform daily activities, including defecation, was associated with a high rate of constipation in healthy volunteers.
There is no doubt that exercise and physical activity have a bene cial impact on the physiology of the colon. Both have been shown to accelerate gastrointestinal transit, [23,24] increase colon motility, and even produce favorable hormonal changes, thus preventing constipation. [10,25] Nevertheless, it is important to recognize that there are other determining factors, such as the type and intensity of physical activity, as well as water and ber intake. In that context, we found that 15% of the subjects with "newonset" constipation stopped doing exercise during the lockdown (68% before vs 53% after), whereas the rest of the subjects with constipation symptoms that did do exercise, reduced the frequency and intensity of their physical activity.
Another interesting nding of our study was that a signi cant number of subjects that developed constipation drank less than 1 liter of water per day. That result was similarly described in a population study conducted in the United States by Markland et al. [12] They reported that low liquid intake (<1,882 ml per day) was a predictor for constipation in women (prevalence odds ratio (POR): 1.3, 95% CI 1.0,1.6) and men (POR: 2.4, 95% CI 1.5,3.9). The mechanisms by which appropriate liquid intake positively in uences colon transit are variable and include a better general hydration status and it can potentiate the effect of ber and reduce its secondary effects. [26,27,28] With respect to ber, as in other studies, [12] we found no relation between ber intake and the appearance of "new-onset" constipation symptoms. However, we did not utilize a speci c tool for ber intake, unlike other population studies, which could be a source of bias in our study. Regarding the quality of life of the subjects that developed constipation, were found that the mean scores in the subscales and overall score of the PAC-QOL re ected a negative impact on the quality of life of those patients, even in the short span of symptom progression.
Even though ours is a novel study with relevant ndings, it also has limitations. The appearance of "newonset" constipation symptoms does not necessarily imply that they will persist chronically and classi ed as chronic constipation according to Rome IV criteria, and in some subjects, the symptoms will probably disappear once their pre-lockdown physical activity and habits are re-established. Our study also was focused primarily on evaluating constipation symptoms, without taking abdominal pain into account, making it likely that some of our subjects actually had IBS-C. With respect to the tools utilized, the quantity of water was evaluated through an arbitrarily created Likert scale, considering that normal water intake should be at least 1 liter per day. Likewise, the null association between ber intake and constipation was related to the fact that we did not employ a speci c questionnaire for its appropriate evaluation. Finally, psychological comorbidities were not accounted for as these have been shown to have increased during COVID and may have acted as confounders.
In conclusion, in our study on an open population in Mexico, we found that one-fourth of the population that reduced their physical activity and drank less water due to mobility restrictions imposed for epidemiologic reasons, in the face of the COVID-19 pandemic, developed "new-onset" constipation symptoms. Given those results, appropriate physical activity and adequate liquid intake during prolonged periods of lockdown should be recommended to prevent said symptoms. Figure 1 Incidence of constipation during the lockdown Figure 2 Number of bowel movements, days of physical activity and stool consistency before and after the lockdown