The Bowel Research Collaborative Data Tracking System Implementation Study

Site assessment: Each site will be assessed as to their current bowel assessment and documentation methods, and their current bowel management strategy. Clinical practice leaders (nursing and physician) will be interviewed, and as many of the nurses as can be contacted will be asked to complete a brief questionnaire about bowel care at their site, with some focus groups being held in order to allow all staff to have input.

A retrospective chart review will be carried out to include at least 30 patients to establish the current documentation standard and outcomes.

The impact of bowel care on nursing practice will be assessed.

Intervention: As a result of the assessments at all the sites and in collaboration with the site leaders, the research team will devise a common reporting and documentation system which will enable systematic collection of bowel outcomes data. It is intended that the Victoria Bowel Performance Scale will be introduced, in combination with other reporting tools specific to bowel management. The tools may vary from site to site depending on the context of care, for example an inpatient unit will have differing requirements to an outpatient clinic. However the goal is for required outcome data to be retrievable from charts from all settings in a consistent, complete and reliable way.

Evaluation: Medical and nursing staff will be surveyed in the same way as in the initial assessment (interviews, questionnaires and focus groups), to assess the performance and acceptability of the modified reporting process. A second chart review will be undertaken when at least 30 patients have passed through the service. Outcomes from the chart reviews will be communicated back to care providers in order to give them feedback as to the quality of bowel care they are providing. Any further modifications to the data tracking system will be made so that by the end of the project the system is acceptable to all sites and functioning effectively, allowing the network of collaborating sites to proceed to the next step: clinical trials of different bowel care treatments.

Demographic data recorded will be that required to determine comparability of cohorts with respect to risk for constipation and will include the following variables.

• Age
• Gender
• Palliative Performance Status (PPSv2) on admission
• Diagnosis (if cancer, which tumour type)
• Major comorbid diagnoses
• Reason for admission
• Presence or absence of abdominal cancer
• Presence or absence of ascites
• Radiotherapy during admission (if yes, to abdomen or not)
• Medications taken during admission
• Duration of admission
• Number of days on bowel protocol
• Discharge or death

The primary outcome measures for bowel function for each patient will be the proportion of days with at least one bowel movement a day, and the proportion of patients with a bowel movement at least 40 % and at least 50% of days. The secondary outcome variables of the study will be use of enemas, suppositories or lactulose, and reported cramping and/or diarrhoea. Diarrhoea will be defined as either passage of watery stool at least once a day, or more than 3 bowel movements a day. These were the same outcome variables used in the docusate study. Sites will be grouped according to type of setting (hospice, PCU, home etc). Sites with very similar patients will be combined for statistical analysis (for example the two Vancouver residential hospices to which patients are referred through a common process and assigned to according to bed availability).

Acceptability of the BPS will be assessed through a combination of questionnaires and focus groups with staff at sites which elect to pilot it after reviewing the results of the initial assessment. Their responses will be analysed using qualitative methodology. Staff will also be asked at the end of the study whether or not they would prefer to continue to use the BPS for routine clinical assessment or to revert to their previous method.

The utility of the BPS as a suitable outcome measure for clinical trials of bowel management treatments will be explored by correlating the single item BPS score with the primary and secondary outcome variables. As scoring of the BPS is essentially a composite of multiple outcome variables, (bowel movement frequency, consistency, degree of sphincter control and associated symptoms such as cramping etc.) it is anticipated that the BPS will provide at least the same outcome information as the multiple individual variables, which are rarely fully documented. We hope that the BPS will provide more patient-centred and useful information to guide laxative treatment, as the tool is centred around the individual's "normal" bowel habit, rather than relying on arbitrarily defined norms. Data from all sites will be able to be pooled for this analysis.