Healthcare resource use and costs of opioid-induced constipation among non-cancer and cancer patients on opioid therapy: A nationwide register-based cohort study in Denmark

1.1 Background

Opioids are frequently used for chronic pain but they are not often effective [1]. The majority of opioid users in Denmark are individuals with acute pain, while the highest consumption is attributable to users with non-malignant chronic pain [1]. A well-known and common side effect of opioid use is bowel dysfunction which can lead to constipation. The frequency of opioid-induced constipation (OIC) ranges from 33 to 94% in non-cancer and cancer patients [2,3,4]. Results from a population-based survey of 322 patients with chronic pain in the US and EU, showed that 33% of the patients reported that they had missed doses, decreased the dose of or stopped using opioid medication in order to relieve bowel-related side effects [5,6].

OIC may lead to opioid dose reduction, non-adherence, and treatment discontinuation resulting in possible inadequate analgesia [6]. In addition, persistent constipation may lead to serious medical conditions, such as bowel obstruction and faecal impaction, and consequently to increased use of healthcare services, decreased work productivity, and reduced quality of life [7,8,9,10].

Previous studies have shown that OIC imposes a substantial economic burden for patients, however without differentiating between the underlying indications for opioid treatment, such as cancer versus non-cancer pain [7,11,12,13]; and only a few studies have evaluated constipation after opioid therapy in patients with non-cancer pain [14,15]. There is thus a substantial need for population-based studies assessing the economic and health-related burden of OIC, focusing on differentiating between patients with non-cancer and cancer pain.

1.2 Objectives

The objectives of this nationwide register-based cohort study was to determine healthcare resource use and cost among non-cancer and cancer patients with OIC, with a focus on treatment and general management of opioid therapy and OIC in Denmark.

2.1 Study design

This is an observational cohort study on opioid-induced constipation (OIC) identified from national health registries in Denmark during 1998–2012.

2.2 Setting

Denmark have a tax funded health care system providing all inhabitants free healthcare access (including both general practices and hospitals) irrespective of socioeconomic status [16].All Danish citizens are registered in the Danish civil Registration system and assigned a unique civil registration number, which is utilized in all national registers thus enabling accurate register linkage. In this study, healthcare resource utilization and costs included hospitalizations, outpatient visits, drugs, and visits to general practitioners and practising specialists in the public and private health care sector in Denmark. The following data sources were utilized:

• The Danish National Patient Registry [17], including data on all somatic hospitalizations (since 1977) and outpatient visits (including emergency room visits) since 1995. Hospital discharge and outpatient contact diagnoses are according to the International Classification of Diseases (ICD version 10 from 1994 onwards).

• Cost of hospitalization and outpatient cost were based on diagnosis-related group tariffs from the Danish Ministry of Health.

• The use and costs of drugs were based on data from the National Danish Medicine Agency, which includes the retail price of the drug (including dispensing costs) multiplied by the number of transactions. The Danish National Prescription Registry [18] keeps records on all drug prescriptions issued from Danish pharmacies since 1995. Due to partial reimbursement of drug expenses by the Danish health care system, all pharmacies in Denmark are required to register each prescription issue in the national prescription registry, ensuring complete registration. Each issued drug prescription is registered according to the Anatomical Therapeutic Chemical (ATC) system [19].

• The frequency and costs of consultations with general practitioners and other specialists were based on data from the National Health Security.

• Data on socioeconomic status was retrieved from the Danish Income Statistics. Classification of education and employment status were performed on individual level (Table 1).

2.3 Study population

All patients aged 18 years or older who had issued a prescription for opioids, with a minimum total daily dose of 30 mg of oral morphine, or equianalgesic dosing of one or more opioid therapies (ATC: N02 [analgesics] except N07BC [drugs used in opioid dependence] and N02AC [diphenylpropylamine derivatives]) during the study inclusion period 01.01.1998 to 30.06.2012 were identified in the Danish National Prescription Registry.

Eligible patients had: (1) treatment with opioids for at least four weeks and at least two prescriptions issued with known opioid coverage of at least 70% (defined as number of daily doses dispensed divided by the total number of days between prescription issue); (2) at least 12 months opioid free period prior to the index date (defined as the date of first of at least two opioid prescriptions issued); (3) at least 6 months of follow-up after index date; (4) no cancer diagnosis (ICD10: C00-C99) after study entry (due to the complexity of symptoms prior to a cancer diagnosis as well as the long study period); (5) no missing information on socioeconomic status or duration of opioid treatment.

2.4 Variables

3 Results

Approximately 1.4 million patients over 18 years of age initiated opioid therapy during January 1998 to June 2012 and 119 922 of these patients were in treatment with opioids for ≥4 week and had ≥2 prescriptions issued with known opioid coverage of ≥70% (Fig.1).For the current analyses, the following exclusions were made: 9814 patients with <6 months of follow-up; 8457 patients diagnosed with cancer during the study period; 4482 patients missing information on socioeconomic status and opioid duration.

Fig.1

Flow chart of cohort selection and sample size (abbreviations: OIC = opioid-induced constipation, n = number).

Of the 97 169 patients eligible of analyses 77 568 patients were non-cancer patients and 19 601 patients had a cancer diagnosis prior to study entry. Among the non-cancer patients,15% were classified with OIC, 10% with previous constipation, and 75% without OIC, while for the cancer patients the corresponding figures were 35%, 14%, and 51%, respectively.

3.1 Non-cancer patients

Mean age was 61 years (standard deviation [SD] 18), 73 years (SD 16), and 75 years (SD 16) for the non-cancer patients without OIC, with OIC, and with previous constipation, respectively (Table 1). Irrespective of which OIC status definition was met, the majority were women (Table 1).A higher proportion of non-cancer patients with OIC were periodic (41%) or strong opioid users (69%), compared to those without OIC (31% and 41%, respectively). Further, the mean number of days on opioids were higher for patients with OIC (1189 days [SD 1164])than for both without OIC (584 days [SD 898]) or with previous constipation (573 days [SD 782]). The overall most prevalent co-morbidity was cardiovascular disease, more commonly seen for the patients with current or previous constipation (31% and 36%, respectively), than for non-cancer patients without OIC (19%) (Table 1).Primary school was the most prevalent educational level in all three OIC status definitions (Table 1).In the OIC and previous constipation patient groups, more than 70% of patients were retired due to age compared to only about 40% of the no OIC patient group.

There were approximately four hospital admissions per patient year in all three constipation status definitions for the non-cancer patients (Table 2). The length of hospital stay was about 7 days per patient year for patients with current or previous constipation and about 4 days per patient year for patients without OIC. Among non-cancer patients with OIC and previous constipation, the absolute total healthcare cost were 9654 Euros and 10 249 Euros per patient year, respectively. Among non-cancer patients without OIC the total healthcare cost were 6986 Euros per patient year. In all three OIC status definitions, total costs were mostly driven by costs from inpatient visits and prescribed medication costs.

The multivariable regression analysis (Table 3) shows that noncancer patients with OIC showed 34% higher costs compared to those without OIC (P < 0.001), whereas costs were 20% higher in non-cancer patients with previous constipation than in those without OIC (P < 0.001), after adjusting for age, gender, opioid usage, marital status, and comorbidity. The use of strong opioids and short-term opioid treatment were also important cost drivers among non-cancer patients. Strong opioid users had 76% higher costs compared to weak opioid users, whereas short-term opioid users (1–3 months and 4–6 months) had more than two-fold higher costs compared to long-term opioid users (above 12 months).

3.2 Cancer patients

The mean age among cancer patients varied from 67 years (SD 12) in the no OIC and OIC groups to 71 years (SD 13) in the previous constipation group (Table 1a – Appendix A). A minority were women in all three constipation status definitions (Table 1a). A higher proportion of cancer patients with OIC were continuous opioid users (85%) and strong opioid users (97%), compared to cancer patients without OIC (83% and 85%, respectively). Further, the mean number of days on opioids were higher for cancer patients with OIC (329 days [SD 523]) than without OIC (238 days [SD 447]). For all three constipation status definitions, cardiovascular disease was the most prevalent co-morbidity and primary school was the most prevalent educational level. In the previous constipation group, approximately three out of four patients were retired due to age compared to about half of the cancer patients with OIC and without OIC.

OIC and previous constipation cancer patients, had 21 and 20 hospital admissions per patient year, respectively. Cancer patients without OIC had 17 hospital admissions per patient year (Table 2). The length of hospital stay was 23 days per patient year for patients with current or previous constipation, and 18 days per patient year for patients without OIC. The absolute total healthcare costs among cancer patients with current OIC, previous constipation, and without OIC were: 32 060 Euros, 29 961 Euros, 24 925 Euros per patient year, respectively.

The results from the multivariable regression (Table 3) shows that cancer patients with OIC had 25% higher costs compared to patients without OIC (P < 0.001), whereas costs were 7% higher in patients with previous constipation than in those without OIC (P < 0.001) after accounting of age, gender, opioid usage, marital status, and comorbidity. The use of strong opioids and short-term opioid treatment were also important cost drivers among cancer patients, showing an 81% higher cost among strong compared to weak opioid users and a two-fold higher cost among short-term versus long-term opioid users.

4 Discussion

5 Conclusion

The results of this nationwide study based on real life data suggested that patients with OIC, regardless of cancer or non-cancer status, have higher healthcare resource utilization and higher associated costs compared to patients without OIC. The total healthcare costs were mostly driven by costs from inpatient visits and prescribed medications.

Implications

Reducing the number of OIC patients has potential cost savings for the health care system. Special attention should be given to patient groups who might be at high risk of opioid-induced constipation, such as users of strong opioids, patients on long-term opioid treatment, at advanced age, and with concomitant cardiovascular disease.