This is an observational descriptive study using a commercially available hospital claims database provided by Medical Data Vision Co., Ltd. (MDV) [19]. The database contains over 39 million patient demographics (including age and sex), ICD-10 coded diagnoses, medical practices, hospitalizations, and prescribed medications from over 460 hospitals, which covers 26% of Japanese hospitals that use an acute inpatient care system called the Diagnostic Procedure Combination [19]. MDV collects the administrative data that are anonymized at the hospitals, receives the consent of the hospital, and processes them into the database.

Ethical approval and consent were not required according to Ethical Guidelines for Medical and Health Research Involving Human Subjects of the Ministry of Health, Labour and Welfare, Japan, because the data had been anonymized.

From the MDV database, eligible patients were identified on the basis of the following criteria: (i) diagnosis of cancer (ICD-10 codes C00-D09), (ii) initiation of opioid analgesic therapy after the cancer diagnosis, and (iii) long-term opioid prescription (continuous prescriptions for 30 days or more of any of the following opioids: morphine, hydromorphone, oxycodone, fentanyl, tapentadol, methadone, codeine, tramadol, pentazocine, and/or buprenorphine). The continuous prescriptions were defined if the blank periods of opioid prescription were less than 2 days. Opioids designed for rescue therapy and buprenorphine transdermal patch, which is not indicated for cancer pain in Japan, were not included in the initiation of opioid analgesic therapy.

Patients who received laxatives within 3 months or opioids within 7 days before initiating opioid analgesic therapy were excluded. The data period was the initiation of opioid analgesic therapy and was between January 2018 and December 2019. It examined the use of naldemedine, a PAMORA that was launched in Japan in 2017, and avoided the effects of the COVID-19 lockdown, which began in April 2020 in Japan. Each patient was followed for 3 months after initiating opioid analgesic therapy.

All data were retrospectively obtained from the database. Age, sex, cancer type, outpatient or inpatient status, and opioids at initiation were collected. Codeine, tramadol, and pentazocine were defined as weak opioids, and other opioids were defined as strong opioids (Table S1 in the Supplementary Material) on the basis of the Clinical Guidelines for Cancer Pain Management [14]. The doses of opioids were not considered.

Treatment patterns of laxatives for OIC were investigated in both groups: the early medication initiation group, which received laxatives within 3 days, including the same day after initiating opioid analgesic therapy, and the non-early medication group (Fig. 1). Laxatives were classified into four categories: osmotic laxatives, stimulant laxatives, PAMORAs, and others. The categories of individual laxatives are provided in Table S2 in the Supplementary Material. The first-line laxatives for each opioid type were examined on the basis of the class and route of administration (oral weak opioids, oral strong opioids, and transdermal strong opioids). The proportion of laxatives was calculated as the percentage of the total number of patients who started each opioid type.

A scheme of study design. Patients with cancer who initiated continuous opioid analgesic therapy (≥ 30 days) were included. Early and non-early laxative medication was defined when laxative prescription was initiated within 3 days or not after initiating opioid analgesic therapy, respectively

In addition, the following were evaluated for each first-line laxative: number of days of first-line laxative medications, duration from the initiation of opioid analgesic therapy to the start of first-line laxative prescription, change or addition of new laxatives, dose increase, number of laxatives used, lost to follow-up, and second-line laxatives. All data aggregations were carried out with SAS v.9.4 (SAS Institute Inc., Cary, NC, USA).