Clinical Implementation of a Pictographic Constipation Action Plan for Children With Functional Constipation

Bakcground

Worldwide, it is estimated that constipation affects 0.7% to 29.6% of children. Constipation accounts for 3% to 10% of general pediatric outpatient visits and up to 25% of visits to pediatric gastroenterologists. Constipation is often associated with infrequent and/or painful defecation, fecal incontinence, and abdominal pain. Furthermore, it may result in significant distress to the child and family, and has a significant impact on health care cost, school absenteeism, and parental workdays missed. While there are concerning organic causes of constipation, - the diagnosis for the 90% of pediatric constipation complaints is not attributable to an organic etiology and is thus classified as functional.

When a child meets these criteria, the diagnosis of functional constipation can be made when, after appropriate evaluation is performed, the patient's symptoms cannot be fully explained by another medical condition.

Multiple research investigations have demonstrated the negative effects on quality of life of pediatric patients with functional constipation as well as the negative effect of this diagnosis on their parents and other caregivers. This impaired health related quality of life (HRQoL) exhibited by children with constipation is lower not only than those of healthy controls, but also lower than children with other significant gastrointestinal diseases such as inflammatory bowel disease or gastroesophageal reflux disease.

In 2014, the pediatric gastroenterology team from Boston Children's Hospital began to advocate for a new "Shared Care" model which implemented of a collaborative quality improvement plan by coordinating a special education program for surrounding primary care managers (PCM) with a focus on the diagnosis and management of functional constipation in children. After developing a constipation management algorithm, The Boston Children's team would conduct digital education with PCMs to discuss constipation pathophysiology as well as the application of the Boston Children's constipation management algorithm. Post-intervention analysis revealed decreased referrals to the pediatric gastroenterology service, which was attributed to improved provider education.

Beyond empowering providers in the chronic care of children, there are multiple examples of other novel interventions to empower patients and parents to better manage chronic conditions, when away from the Patient Centered Medical Home.

Specifically, clinical symptom action plans including those for constipation, asthma, anaphylaxis, and eczema have all been used in the clinical arena. These plans are tailored specifically to the medical needs of the have demonstrated the ability to decrease hospital admissions, decrease emergency department visits, and improve satisfaction for home care by parents. Of note, investigators at the Albert Einstein Colle of Medicine have shown that the implementation of an asthma action plan program decreases hospital admissions, decreases emergency department visits, and decreases medical resource cost.

While appropriate pharmacotherapy and the use of constipation action plans have demonstrated varying scientific rigor within the literature, current clinical practice guidelines and available evidence fail to address the root of the constipation problem in children- which include lifestyle factors such as diets low in fiber, inadequate hydration, poor posturing and other maladaptive behaviors. Direct patient education is key to help address and overcome these obstacles. Most recently a clinical questionnaire for children has placed an emphasis on the clinical impact of behaviors and posturing with the release of the Rome IV Questionnaire for Pediatric Functional gastrointestinal Disorders.

Prior to this, however, current clinical practice guidelines and action plans for constipation have failed to address what many gastroenterologists would consider the root of the constipation problem- poor posturing. The effect of inappropriate straight posturing on the puborectalis angle and the detrimental effect it can have on defecation. The puborectalis muscles acts like a sling encircling the lower colon. When the back is straight and the knees are below the level of the abdomen, the puborectalis muscle contracts causing the colon to kink, thereby impeding the evacuation of stool. When the angle is changed by lifting the knees higher, an angle of 35 degrees can be attained which has been associated with decreased constipation in adults. As a standard of care, posture modification has been made possible through the use of a Defecation Posture Modification Device (DPMD). As a minimal-cost, non-pharmacologic means of therapy, the DPMD is postured to enact real change on the care of constipation in children by serving as a stool for children to prop their feet on and achieve the desired 35 degree angle. This along with other behavioral modifications included in the standard of care: sticker charts, scheduled toilet time, and rewards systems may capture the psychological aspect of functional constipation management which warrants additional investigation.

In summary, investigators seek to create a program which combines improved patient/ provider education, incorporates constipation action plans, to shift the care paradigm to rely more intentionally on written clinical action plans to address low health literacy and improve outcomes.

Methods Investigators measured clinical and patient-related outcomes during this quality improvement project.

At the time of enrollment, participants (or a parent proxy) responded to a health literacy questionnaire: The Newest Vital Sign. This enrollment survey also offered use of a defecation posture modification device (DPMD) to participants at no cost. The remainder of the enrollment survey gathered baseline measurements using validated instruments developed for clinical, patient-related, and health confidence outcomes which would be trended over time to grade USCAP effectiveness. These instruments are discussed in greater detail below. Four months later, participants responded to a second survey which repeated these clinical, patient-related, and health confidence measures.

The 4-month study period was selected based on our institutional practice pattern and to capture recurrences in constipation that can occur after the discontinuation of pharmacotherapy following shorter treatment durations outlined in the North American Society for Pediatric Gastroenterology Hepatology and Nutrition (NASPGHAN) clinical practice guideline. Survey 2 used common, real-world, clinically relevant questions to assess for pharmacotherapy adherence. Investigators used two questions adapted from the Self-Report Measure for Medication Adherence from Wilson et al. In order to assess adherence both qualitatively and quantitatively, participants were asked, "How often does the patient take his/her constipation medicine in the way he/she is supposed to?" using the 5-point Likert scale (0- Never, 1- Almost Never, 2-Sometimes, 3-Often, 4-Almost Always). Next, participants responded to the question, "How many months did the patient take their constipation medicines for?" using a continuous rating from 0 to 4 months. Finally, participants were asked if they used a DPMD with every bowel movement.

Measurements of Implementation Effectiveness Investigators evaluated the effectiveness of the USCAP through multiple approaches. These include clinical, adherence, patient-related, and health confidence outcomes. Toilet-trained subjects, aged ≥4 years, who met the ROME IV criteria for FC were block randomized to receive either the USCAP or control. All discharge instructions were to continue pharmacotherapy for 4 months. Investigators measured subjects at enrollment and 4 months. Metrics included: clinical outcomes using the ROME IV Pediatric Diagnostic Questionnaire, patient-related outcomes using the PedsQL™ Gastrointestinal Symptoms Module [PedsGIQL] (a 5-point Likert scale: 0- Never, 1- Almost Never, 2- Sometimes, 3-Often, 4-Almost Always). The PedsGIQL was administered to participants on at the start of the project after the clinic visit concluded, and 4 months after initial visit. The goal was to demonstrate an improvement in HRQoL. HRQoL responses were stratified by calculating Total Scale Scores. Higher Total Scale Scores demonstrated less (fewer) gastrointestinal symptoms and thus higher (better) gastrointestinal-specific HRQoL. Total Scale Scores were calculated by taking the sum of the reverse-scored items and then dividing by the number of items responded to. Finally, self-efficacy using the Health Confidence Score [HCS] (4-point Likert scale: 3-strongly agree, 2-agree, 1-neutral, 0-disagree).

DATA ANALYSIS Treatment success rates of the treatment and control groups (i.e., the proportion of each group which no longer met the definition of functional constipation) were compared at project conclusion. Investigators also performed an analysis to determine the proportion of subjects with self/parent-proxy reported Pediatric Bristol Stool Form Scale (PBSFS) 3 or 4 achieved at the close of the project and the changes in rates of fecal incontinence (FI). Investigators compared Total Scale Scores from the PedsGIQL between treatment and control groups to determine whether significant change occurred during the project. Changes in HCS were analyzed to compare whether the USCAP is associated with higher, sustained health-confidence compared to control. Univariate analyses were conducted using the Kolmogorov-Smirnov test to investigate potential outliers and normality of the data. When continuous variables were found to be normally distributed, values were summarized using means and standard deviations (SD). When data were found to be skewed or non-normal data were summarized with medians and interquartile ranges (IQR). When comparing continuous variables between groups either a Student's t-test or Wilcoxon rank sum tests were used. Frequency data were reported as counts and percentages. Chi-square was used for all categorical comparisons. When any condition had N ≤5, a Fisher's Exact Test was applied. A P-value <0.05 was considered statistically significant. SAS 9.4 (SAS Institute, Cary, NC) was used for all analyses.