ORIGINAL RESEARCH published: 21 October 2021 doi: 10.3389/fphar.2021.750193 Geographical, Socioeconomic, and Gender Inequalities in Opioid Use in Catalonia Sara Serra-Pujadas 1†, Cristina Alonso-Buxadé 1†, Júlia Serra-Colomer 1†, Júlia Folguera 1†, Neus Carrilero 2* and Anna García-Altés 2,3,4 1 Facultat d’Economia i Empresa, Universitat Pompeu Fabra, Barcelona, Spain, 2Agència de Qualitat i Avaluació Sanitàries de Catalunya (AQuAS), Barcelona, Spain, 3CIBER de Epidemiología y Salud Pública (CIBERESP), Barcelona, Spain, 4Institut d’Investigació Biomèdica (IIB Sant Pau), Barcelona, Spain Background: In recent years, worldwide opioid use has seen a sharp increase, especially for the treatment of chronic non-cancer pain. Catalonia is no exception to this trend. However, no recent studies have addressed the socioeconomic and gender inequalities in opioid use in the different geographical areas of Catalonia. Methods: We performed an ecological study to analyse the associations between Edited by: socioeconomic status, gender and the use of opioids in the 372 Health Areas of Anick Bérard, Catalonia. Robust Poisson models were performed to analyse the data provided from Université de Montréal, Canada the Central Register of Insured Persons and dispensing data from the Electronic Reviewed by: Prescription Database. Odile Sheehy, University of Montreal, Canada Results: The results show that socioeconomic status has a major impact on opioid use, with Jinping Zhao, University of Montreal, Canada the most deprived areas presenting the highest levels of use. There are major inequalities in the DDD/1,000 inhabitants per areas. Moreover, women have much higher utilization rates than *Correspondence: Neus Carrilero men, especially in more deprived areas. The greatest difference is observed in the use of weak neus.carrilero@gencat.cat opioids in women: the DDD/1,000 inhabitants per day was 2.34 in the area with the lowest use, † These authors have contributed compared with 22.18 in the area with the highest use. equally to this work and share first authorship Conclusions: Our findings suggest that stronger action is needed to promote best practices in prescribing for chronic pain and to reduce socioeconomic and gender variation between Specialty section: This article was submitted to geographical areas. This study provides a model for routine monitoring of opioid prescription for Pharmacoepidemiology, targeted interventions aimed at lowering high-dose consumption in specifically identified areas. a section of the journal Frontiers in Pharmacology Keywords: opioids, gender, socioeconomic factors, prescriptions, medical bias, chronic pain Received: 30 July 2021 Accepted: 24 September 2021 INTRODUCTION Published: 21 October 2021 Citation: Pain is a growing public health priority. Demographic changes, rising comorbidity rates, the Serra-Pujadas S, Alonso-Buxadé C, difficulty of making accurate diagnoses, and poor access to treatment are some of the challenges Serra-Colomer J, Folguera J, faced by healthcare systems in the management of chronic pain. Among all the treatment options, Carrilero N and García-Altés A (2021) opioids have played an increasingly important role in the control of moderate to severe pain Geographical, Socioeconomic, and Gender Inequalities in Opioid Use (Goldberg and McGee, 2011; Majors-Foley, 2016; Andrews et al., 2018). in Catalonia. The main indication of strong opioids is for acute pain of moderate to severe intensity, especially for Front. Pharmacol. 12:750193. oncological or postoperative pain (World Health Organisation, 1996). Due to their known side effects, they doi: 10.3389/fphar.2021.750193 must be monitored under the strict control of a healthcare professional (Labianca et al., 2012). Recent data Frontiers in Pharmacology | www.frontiersin.org 1 October 2021 | Volume 12 | Article 750193 Serra-Pujadas et al. Opioids use Inequalities in Catalonia have shown an increase in opioid consumption for the treatment of Data Sources chronic non-cancer pain (CNCP), including osteoarthritis of the Two data sources were used. The first was the Central extremities and spondylarthritis as the most common pathologies Registry of Insured Persons, in order to obtain the (Agència de Qualitat i Avaluació Sanitàries de Catalunya, 2018). reference population for 2019. This is an automated file Although opioids are useful for alleviating pain, their use and side that includes all residents in Catalonia and their effects have not been thoroughly studied in the case of many sociodemographic characteristics (sex, age, HA of pathologies that cause CNCP, and little evidence has been reported residence). The second was the Electronic Prescription on the risk-benefit balance in these cases. Even so, many clinical Database, an administrative database providing data on practice guidelines maintain that these drugs should be second-line outpatient dispensing, corresponding to the drugs drugs (that is, administered under strict monitoring by a healthcare dispensed by all pharmacy offices during the year 2019. It professional) due to their negative effects on cognition and their has been systematically validated by the healthcare system, addictive potential (Kahan et al., 2011; Labianca et al., 2012; which pays part of the prescription, and by the providers Manchikanti et al., 2012; Scottish Intercollegiate Guidelines (i.e., the pharmacies) which carry out the billing. From this Network (SIGN), 2013; Agència de Qualitat i Avaluació Sanitàries database, we obtained all the doses dispensed in defined daily de Catalunya, 2018). doses (DDD) for each of the opioids selected according to age, Given the generalised increase in consumption, various indicators sex, HA and medical diagnosis associated with the of misuse (including addiction rates or overdose-related deaths) have prescription. Drugs were all coded according to been described in different contexts, and have revealed similar trends Anatomical Therapeutic Chemical (ATC) system (level 7) (Kalkman et al., 2019; Schieber et al., 2019). This trend has been and medical diagnoses according to the Clinical Classification observed on a large scale in the United States, the country with the Software (CCS) level (HCUP CCS, 2017). most opioid users in the world: between 1999 and 2007, sales of opioid drugs increased by 149% (Manchikanti et al., 2013), and there were Variables 399,230 overdose deaths (Scholl et al., 2018). As a result, in 2017 the Dependent Variable US declared this issue a public health emergency and opioid abuse is The study included all opioid drugs authorised for analgesic still considered an epidemic today (US Department of Health and purposes and available at community pharmacies in Spain in Human Services, n.d.). 2019. We distinguished between strong and weak opioids Though less pronounced than in the United States, a similar following the classification recorded at the 1961 Convention of trend has been observed in Europe in the last few years (Bosetti Narcotic Drugs (Lande, 1962). Drugs were assigned to each et al., 2019; European Monitoring Centre for Drugs and Drug category depending on their level of analgesia and Addiction, 2020). Increases in opioid use have also been dependency. Then, we listed the opioids as strong or weak recorded in Spain, where consumption rose by 88% between according to their ATC7 codes: strong, morphine (N02AA01 2010 and 2018, with a notable increase in the use of strong and N02AA03), oxycodone (N02AA05 and N02AA55), opioids (Agencia Española de Medicamentos y Productos phenylpiperidine derivatives (N02AB03 and N02AB02), Sanitarios, n.d.). The same trend is observed in Catalonia, tapentadol (N02AX06), and buprenorphine (N02AE01); and where total use of opioids between 2012 and 2016 increased weak, codeine and combinations (N02AJ06, N02AJ07, and by 45% (Pando et al., 2017). Among strong opioids, the rise in N02AJ08), and tramadol (N02AJ13, N02AJ14, N02AX02, and fentanyl use was particularly marked. N02AX52). An issue that has not been widely addressed so far is the presence The DDD/1,000 inhabitants per day is the main unit measure of inequalities in opioid use. The analysis of the social determinants of of the study. The DDD is a global standardised measure of drug opioid use has been paid far less attention than clinical characteristics consumption that corresponds to the assumed average and age as the main determinants of its increase, despite the reports of maintenance dose per day for a drug used in its main ethnic, socioeconomic, gender, and geographical variations (Morden indication among adults. DDDs are established and updated et al., 2014; Song, 2017; Santoro and Santoro, 2018; Friedman et al., by the WHO (WHO Collaborating Centre for Drug Statistics, 2019; Schieber et al., 2019; Park and Wu, 2020). n. d.). The DDD/1,000 inhabitants per day measure provides a The aim of the present study is to analyse the geographical, rough estimate of the proportion of the study population treated socioeconomic, and gender determinants of the use of opioids daily with a particular drug or group of drugs. for CNCP in Catalonia. Independent Variables Socioeconomic variables: The composite socioeconomic index (ISC, MATERIALS AND METHODS Catalan acronym) was used to measure the socioeconomic level of each HA, and specifically the level of deprivation. The ISC includes Design and Study Population mainly employment (unemployed rate and occupational status), social Multi-group ecological study using the health area (HA) as (educational level), economic (income level), health (hospitalization unit of analysis. In 2019, the Administrative Health Division rate and preventable mortality) variables. It has shown a high divided Catalonia into 372 HAs. The entire Catalan correlation with healthcare use and morbidity in HA (Colls et al., population as of 2019 was considered as the study 2020). Higher values on a range from 0 to 100 indicate higher levels of population, comprising a total of 7,653,845 inhabitants. deprivation). We grouped the HAs into quartiles according to their Frontiers in Pharmacology | www.frontiersin.org 2 October 2021 | Volume 12 | Article 750193 Serra-Pujadas et al. Opioids use Inequalities in Catalonia ISC: Q1 (<34), Q2 (35–42), Q3 (43–50) and Q4 (>50). Of the 372 HAs RESULTS in Catalonia, six were created after the last update of the ISC; in these cases, we assigned the ISC of the HA to which they previously Opioid Use in Catalonia belonged. The total study population comprised 3,755,837 (49.07%) men and Sociodemographic variables: Sex, HA, and age of the opioid users, 3,898,009 (50.93%) women. Opioids were used by 218,354 males categorised into five groups: 0–14, 15–44, 45–64, 65–74, and 75 years (36.44%) and 380,904 females (63.56%). (See Table 1 for the main or more. characteristics of the study population). Clinical variable: A clinical binary variable was Table 2 presents the DDD/1,000 inhabitants per day of strong and constructed to reflect whether the opioid was prescribed weak opioids according to age, ISC, and diagnosis. Median use of all for a cancer diagnosis: all cancer diagnoses included in the opioids was 11.66 (Interquartile Range (IQR) 3.61–21.78) DDD/1,000 multi-level category CCS 2 (neoplasm), vs all other patients inhabitants per day; for strong opioids, it was 4.72 (IQR 1.47–12.4) who did not have a cancer diagnosis (CNCP). DDD/1,000 inhabitants per day, and for weak opioids, it was 7.03 (IQR 2.14–15.41). Use of both opioid types was higher in women than Data Analysis in men. For both types of drugs and for both sexes, use rates were First, we performed a descriptive analysis of opioid users of Catalonia higher in HAs with lower ISC. by age, socioeconomic status and sex. Opioid use increased with age, more so among women than Second, we determined the rate of opioid use (DDD/1,000 among men. For instance, the DDD/1,000 inhabitants per day of inhabitants per day) by drug type, sex, socioeconomic status, age strong opioids in women over 75 years old was 177% higher than group, and diagnosis. The use rate was calculated by using the in women between 65 and 74, compared with a corresponding dispensing data to obtain the total DDD according to the increase in men across the same age of approximately 88%. sociodemographic, socioeconomic and clinical variables and the Patients who used opioids to treat CNCP showed higher use reference population as a denominator. We then performed a rates, for both weak and strong types. Again, there was a clear descriptive analysis of the DDD/1,000 inhabitants per day by difference of DDD/1,000 inhabitants per day according to sex, drug type. with higher opioid use in women (see Table 2). Then, we conducted a direct standardization of DDD/ Figure 1 also shows higher use in women than in men, for all 1,000 inhabitants per day by age using the total population kinds of opioids. In both sexes, tramadol was the most used and of Catalonia in 2019 as our reference population, for both morphine the least. sexes. To obtain a summary of all the information, the standardised values were displayed using maps of the HAs Inequalities in Opioid Use in Catalonia for both men and women. Figure 2 shows the distribution of HAs according to the ISC Finally, robust Poisson models were run to test the association quartiles. The most advantaged HAs were located in the city between opioid use (at DDD/1,000 inhabitants per day) and ISC of Barcelona, in the other three main urban regions (Girona, quartiles by age and sex for both weak and strong opioid types. The Tarragona, and Lleida), and in Barcelona’s northern models were adjusted for cancer diagnosis and age. Additional robust metropolitan area. Nonetheless, it was in the metropolitan Poisson models were run, stratified by cancer diagnosis. From these area where the greatest inequalities were observed. models, we estimated the odds ratio (OR) with corresponding 95% Figures 3, 4 display comparative maps of the distribution confidence intervals (CI) and associated p-values. of opioid use in DDD/1,000 inhabitants per day standardised The statistical analysis was carried out using the STATA by age for each HA, based on sex and drug type. The maps 16.1 IC (StataCorp, n.d.) software, and the maps used to show a visual correlation: the lower the ISC, the lower the use display the data were designed through the Instamaps app of of opioids. This was particularly clear in the HAs of the city the Cartographical and Geological Institute of Catalonia of Barcelona. In women, the use of weak opioids tended to be (Institut Cartografic Generalitat Catalunya, n.d.). higher in the HAs of Lleida and Tarragona, while that of FIGURE 1 | Opioid use (DDD/1,000 inhabitants per day) by sex and type of drug in Catalonia, 2019. Frontiers in Pharmacology | www.frontiersin.org 3 October 2021 | Volume 12 | Article 750193 Serra-Pujadas et al. Opioids use Inequalities in Catalonia TABLE 1 | Descriptive characteristics of the population and opioid users of Catalonia, 2019. All population Opioid users Men Women Men Women N (%) N (%) N (%) N (%) Age (years) 0–14 588,812 15.68 554,721 14.23 281 0.13 345 0.09 15–44 1,456,934 38.79 1,416,080 36.33 49,922 22.86 64,320 16.89 45–64 1,082,265 28.82 1,092,157 28.02 79,648 36.48 123,597 32.45 65–74 342,700 9.12 396,059 10.16 41,660 19.08 77,787 20.42 ≥75 285,126 7.59 438,991 11.26 46,843 21.45 114,855 30.15 ISCa (quartile) Q1 (least deprived) 1,082,845 28.83 1,185,533 30.41 47,969 21.97 91,876 24.12 Q2 927,086 24.68 951,677 24.41 54,828 25.11 93,623 24.58 Q3 840,638 22.38 851,176 21.84 53,147 24.34 89,224 23.42 Q4 (most deprived) 905,268 24.10 909,622 23.34 62,410 28.58 106,181 27.88 Total 3,755,837 49.07 3,898,009 50.93 218,354 36.44 380,904 63.56 a ISC: Composite socioeconomic index. TABLE 2 | Opioid use (DDD/1,000 inhabitants per day) according to composite socioeconomic index, age, sex, and clinical diagnosis according to type of drug in Catalonia, 2019. DDD/1,000 inhabitants per day Weak opioids Strong opioids Men Women Men Women Age (years old) 0–14 0.03 0.04 0.01 0.00 15–44 1.77 2.31 0,61 0.67 45–64 5.39 10.01 3.24 4.66 65–74 10.45 21.68 5.88 11.67 ≥75 15.40 31.22 11.06 32.37 ISCa (quartile) Q1 (least deprived) 3.24 7.02 2.02 5.21 Q2 4.35 9.32 2.60 6.48 Q3 4.90 10.40 2.77 6.89 Q4 (most deprived) 5.23 11,46 2.86 7.24 Cancer diagnosis Yes 0.11 0.13 0.65 0.60 No 4.26 9.24 1.90 5.78 a ISC: Composite socioeconomic index. FIGURE 2 | Composite socioeconomic index (ISC) in quartiles for the health areas in Catalonia, 2019. strong opioids was higher in the Catalan coastal area. In men, the distribution was less homogeneous and no clear visual pattern of use was observed. bracket where weak opioid use in women was 124% higher in the The largest gap in the DDD/1,000 inhabitants per day in the most deprived HAs than in the most affluent ones (Q4, OR  different HAs was observed in the use of weak opioids in women 2.24, 95%CI  2.12, 2.37, p < 0.001). (Figure 3), where the HA with the lowest rate was 2.34 DDD/1,000 Furthermore, a steeper gradient was observed in women inhabitants per day compared to 22.18 in the HA with the highest rate. than in men, for all ages. This gap was particularly noticeable in the 45–64 years age range, for both strong and weak Factors Associated With Opioid Use opioids. For example, use of strong opioids was 106% Tables 3, 4 show the OR obtained from the robust Poisson higher in women in Q4 areas than in their peers in Q1 models. There was a statistically significant negative association areas (Q4, OR  2.06, 95%CI  1.90, 2.23, p < 0.001), between use rates in the most deprived HAs and those of the most while in men the difference was 60% (Q4, OR  1.60, 95% advantaged ones, in so far as the use in the most deprived HAs CI  1.45, 1.75, p < 0.001). However, the socioeconomic was higher. This was seen in all ages, especially in the age 45–64 gradients observed between areas were, in most cases, steeper Frontiers in Pharmacology | www.frontiersin.org 4 October 2021 | Volume 12 | Article 750193 Serra-Pujadas et al. Opioids use Inequalities in Catalonia FIGURE 3 | Weak opioids use (DDD/1,000 inhabitants per day standardized by age) by sex and health area in Catalonia, 2019. FIGURE 4 | Strong opioids use (DDD/1,000 inhabitants per day standardized by age) by sex and health area in Catalonia, 2019. for weak opioids than strong opioids, for all age ranges and As with the main model, a clear socioeconomic gradient for both sexes. across areas was observed for the use of opioids to treat Tables 5, 6 present the results of the OR obtained from the CNCP. This gradient was less steep in prescriptions for models stratified by cancer diagnosis. The results were similar cancer pain control. to those observed in the previous model. Although not all coefficients reached statistical significance, in most cases, the associations were significant for the most disadvantaged HAs. DISCUSSION However, it should be noted that generally the results for cancer-related use were not statistically significant, whereas We assessed the results of this study from two perspectives: first, results for opioids prescribed for other causes did reach focusing on the socioeconomic dimension, and second, focusing on significance. gender. Frontiers in Pharmacology | www.frontiersin.org 5 October 2021 | Volume 12 | Article 750193 Serra-Pujadas et al. Opioids use Inequalities in Catalonia TABLE 3 | Odds ratio (OR) of weak opioids use by sex, age and composite socioeconomic index (quartiles) adjusted for cancer diagnosis and age in Catalonia, 2019. Men Women OR 95% CI p-value OR 95% CI p-value a 0–14 years old nc nc nc nc nc nc 15–44 years old Q1 (least deprived) 1.00 — — 1.00 — — Q2 1.56 (1.34–1.82) <0.001b 1.64 (1.45–1.86) <0.001b Q3 1.70 (1.46–1.98) <0.001b 1.82 (1.61–2.06) <0.001b Q4 (most deprived) 2.03 (1.76–2.34) <0.001b 2.11 (1.87–2.38) <0.001b 45–64 years old Q1 (least deprived) 1.00 — — 1.00 — — Q2 1.51 (1.40–1.64) <0.001b 1.60 (1.51–1.70) <0.001b Q3 1.77 (1.64–1.92) <0.001b 1.90 (1.79–2.01) <0.001b Q4 (most deprived) 2.08 (1.93–2.25) <0.001b 2.24 (2.12–2.37) <0.001b 65–74 years old Q1 (least deprived) 1.00 — — 1.00 — — Q2 1.40 (1.27–1.55) <0.001b 1.48 (1.39–1.58) <0.001b Q3 1.58 (1.43–1.74) <0.001b 1.66 (1.56–1.77) <0.001b Q4 (most deprived) 1.78 (1.62–1.95) <0.001b 1.91 (1.80–2.03) <0.001b ≥75 years old Q1 (least deprived) 1.00 — — 1.00 — — Q2 1.21 (1.11–1.32) <0.001b 1.20 (1.15–1.26) <0.001b Q3 1.35 (1.24–1.47) <0.001b 1.28 (1.22–1.34) <0.001b Q4 (most deprived) 1.41 (1.30–1.54) <0.001b 1.38 (1.31–1.44) <0.001b a The results for the 0–14 age group are not conclusive, due to the low use in this group. b Statistical significance at ≤0.05%. TABLE 4 | Odds ratio (OR) of strong opioids use by sex, age, and composite socioeconomic index (quartiles), adjusted for cancer diagnosis and age in Catalonia, 2019. Men Women OR 95% CI p-value OR 95% CI p-value a 0–14 years old nc nc nc nc nc nc 15–44 years old Q1 (least deprived) 1.00 (1.00–1.00) — 1.00 (1.00–1.00) — Q2 2.57 (2.03–3.26) <0.001b 1.18 (0.96–1.44) 0.1079 Q3 2.50 (1.96–3.18) <0.001b 1.51 (1.24–1.83) <0.001b Q4 (most deprived) 2.37 (1.86–3.01) <0.001b 1.51 (1.25–1.82) <0.001b 45–64 years old Q1 (least deprived) 1.00 (1.00–1.00) — 1.00 (1.00–1.00) — Q2 1.30 (1.18–1.43) <0.001b 1.44 (1.32–1.56) <0.001b Q3 1.47 (1.33–1.62) <0.001b 1.74 (1.61–1.89) <0.001b Q4 (most deprived) 1.60 (1.45–1.75) <0.001b 2.06 (1.90–2.23) <0.001b 65–74 years old Q1 (least deprived) 1.00 (1.00–1.00) — 1.00 (1.00–1.00) — Q2 1.41 (1.24–1.60) <0.001b 1.56 (1.44–1.70) <0.001b Q3 1.65 (1.45–1.87) <0.001b 1.68 (1.54–1.82) <0.001b Q4 (most deprived) 1.56 (1.37–1.77) <0.001b 1.86 (1.71–2.02) <0.001b ≥75 years old Q1 (least deprived) 1.00 (1.00–1.00) — 1.00 (1.00–1.00) — Q2 1.19 (1.08–1.32) <0.001b 1.25 (1.19–1.31) <0.001b Q3 1.28 (1.15–1.41) <0.001b 1.27 (1.21–1.33) <0.001b Q4 (most deprived) 1.34 (1.21–1.48) <0.001b 1.33 (1.26–1.39) <0.001b a The results for the 0–14 age group are not conclusive, due to the low use in this group. b Statistical significance at ≤0.05%. Frontiers in Pharmacology | www.frontiersin.org 6 October 2021 | Volume 12 | Article 750193 Serra-Pujadas et al. Opioids use Inequalities in Catalonia TABLE 5 | Odds ratio (OR) of weak opioids use by sex, age, composite socioeconomic index (quartiles), and cancer diagnosis adjusted for age in Catalonia, 2019. Cancer No cancer Men Women Men Women OR 95% CI p-value OR 95% CI p-value OR 95% CI p-value OR 95% CI p-value 0–14 years olda nc nc nc Nc nc nc nc nc nc nc nc nc 15–44 years old Q1 (least deprived) 1.00 — — 1.00 — — 1.00 — — 1.00 — — Q2 1.48 (0.27–8.02) 0.649 0.98 (0.36–2.65) 0.961 1.56 (1.34–1.82) <0.001b 1.65 (1.46–1.87) <0.001b Q3 1.15 (0.18–7.19) 0.881 1.30 (0.50–3.38) 0.587 1.70 (1.46–1.98) <0.001b 1.83 (1.61–2.07) <0.001b Q4 (most deprived) 1.88 (0.38–9.37) 0.440 1.36 (0.54–3.44) 0.510 2.03 (1.76–2.35) <0.001b 2.12 (1.88–2.40) <0.001b 45–64 years old Q1 (least deprived) 1.00 — — 1.00 — — 1.00 — — 1.00 — — Q2 1.31 (0.77–2.22) 0.324 1.31 (0.83–2.05) 0.249 1.52 (1.40–1.65) <0.001b 1.61 (1.52–1.70) <0.001b Q3 1.35 (0.79–2.33) 0.275 1.61 (1.04–2.52) 0.034b 1.78 (1.64–1.93) <0.001b 1.90 (1.80–2.01) <0.001b Q4 (most deprived) 2.02 (1.23–3.32) 0.005b 1.81 (1.18–2.79) 0.007b 2.08 (1.93–2.25) <0.001b 2.25 (2.13–2.38) <0.001b 65–74 years old Q1 (least deprived) 1.00 — — 1.00 — — 1.00 — — 1.00 — — Q2 1.15 (0.70–1.90) 0.583 1.34 (0.79–2.28) 0.274 1.41 (1.28–1.56) <0.001b 1.49 (1.40–1.58) <0.001b Q3 1.01 (0.60–1.73) 0.959 1.57 (0.93–2.64) 0.090 1.60 (1.45–1.76) <0.001b 1.66 (1.56–1.77) <0.001b Q4 (most deprived) 1.49 (0.92–2.41) 0.109 1.57 (0.93–2.63) 0.089 1.79 (1.63–1.97) <0.001b 1.92 (1.80–2.03) <0.001b ≥75 years old Q1 (least deprived) 1.00 — — 1.00 — — 1.00 — — 1.00 — — Q2 1.24 (0.79–1.96) 0.356 1.16 (0.78–1.73) 0.454 1.21 (1.11–1.32) <0.001b 1.20 (1.15–1.26) <0.001b Q3 1.36 (0.86–2.15) 0.194 1.05 (0.69–1.60) 0.826 1.35 (1.23–1.47) <0.001b 1.28 (1.22–1.35) <0.001b Q4 (most deprived) 1.32 (0.83–2.08) 0.238 1.17 (0.78–1.75) 0.461 1.42 (1.30–1.55) <0.001b 1.38 (1.32–1.45) <0.001b a The results for the 0–14 age group are not conclusive, due to the low use in this group. b Statistical significance at ≤0.05%. TABLE 6 | Odds ratio (OR) of strong opioids use by sex, age, composite socioeconomic index (quartiles), and cancer diagnosis adjusted for age in Catalonia, 2019. Cancer No cancer Men Women Men Women OR 95% CI p-value OR 95% CI p-value OR 95% CI p-value OR 95% CI p-value 0–14 years olda nc nc Nc nc nc nc nc nc nc nc nc nc 15–44 years old Q1 (least deprived) 1.00 — — 1.00 — — 1.00 — — 1.00 — — Q2 9.41 (4.63–19.12) <0.001b 1.15 (0.71–1.86) 0.5820 1.92 (1.48–2.50) <0.001b 1.18 (0.95–1.47) 0.1298 Q3 8.33 (4.06–17.09) <0.001b 1.33 (0.82–2.15) 0.2432 1.94 (1.49–2.53) <0.001b 1.54 (1.25–1.90) <0.001b Q4 (most deprived) 3.43 (1.59–7.43) 0.0018b 1.76 (1.13–2.73) 0.0126b 2.27 (1.76–2.92) <0.001b 1.46 (1.18–1.80) 0.0004b 45–64 years old Q1 (least deprived) 1.00 — — 1.00 — — 1.00 — — 1.00 — — Q2 1.05 (0.88–1.26) 0.5633 1.01 (0.84–1.22) 0.8849 1.42 (1.26–1.60) <0.001b 1.57 (1.43–1.72) <0.001b Q3 1.07 (0.89–1.28) 0.4853 1.02 (0.84–1.24) 0.8102 1.66 (1.48–1.87) <0.001b 1.96 (1.79–2.15) <0.001b Q4 (most deprived) 1,29 (1.08–1.53) 0.0043b 1.46 (1.22–1.74) <0.001a 1.75 (1.56–1.96) <0.001b 2.24 (2.05–2.45) <0.001b 65–74 years old Q1 (least deprived) 1.00 — — 1.00 — — 1.00 — — 1.00 — — Q2 1.17 (0.94–1.46) 0,1601 1.17 (0.92–1.49) 0.1958 1.54 (1.32–1.80) <0.001b 1.63 (1.49–1.78) <0.001b Q3 1.65 (1.34–2.03) <0.001b 1.44 (1.14–1.83) 0.0021b 1.65 (1.41–1.93) <0.001b 1.71 (1.57–1.88) <0.001b Q4 (most deprived) 1.40 (1.13–1.74) 0.0021b 1.43 (1.13–1.81) 0.0025b 1.65 (1.41–1.93) <0.001b 1.93 (1.76–2.11) <0.001b ≥75 years old Q1 (least deprived) 1.00 — — 1.00 — — 1.00 — — 1.00 — — Q2 1.03 (0.82–1.29) 0.8156 0.94 (0.78–1.14) 0.5306 1.23 (1.10–1.38) 0.0003b 1.27 (1.21–1.33) <0.001b Q3 1.25 (1.00–1.56) 0.0545 0.93 (0.76–1.14) 0.4753 1.28 (1.14–1.44) <0.001b 1.29 (1.23–1.36) <0.001b Q4 (most deprived) 1.38 (1.11–1.72) 0.0033b 1.01 (0.83–1.22) 0.9347 1.32 (1.18–1.48) <0.001b 1.35 (1.28–1.41) <0.001b a The results for the 0–14 age group are not conclusive, due to the low use in this group. b Statistical significance at ≤0.05%. Frontiers in Pharmacology | www.frontiersin.org 7 October 2021 | Volume 12 | Article 750193 Serra-Pujadas et al. Opioids use Inequalities in Catalonia This study reveals a strong geographical and socioeconomic the ones with the lowest provision of these services (Goodman- gradient in opioid use in Catalonia. The results showed that the Meza et al., 2021). The studies just mentioned and the present one DDD/1,000 inhabitants per day was 10 times higher in the HA all highlight the importance of the health system as a guarantor of with the highest use than in the one with the lowest. The most equitable access to pain treatment and the use of opioid drugs. striking case was the use of weak opioids in women, where the Another noteworthy finding is that the socioeconomic gradient is lowest rate (2.34 DDD/1,000 inhabitants per day) was recorded in more pronounced in intermediate ages, especially in the the Barcelona 5-C HA (an affluent district of the city), and the 45–64 years range, where in many cases the increase in use highest rate (22.18) was recorded in the Badalona 5 HA (a between the most and the least favoured areas exceeds 100%. deprived area). In addition, the descriptive maps show that in This study found a more pronounced socioeconomic gradient metropolitan areas the inequalities were more evident, both in in the prescription of weak than of strong opioids. This may be terms of socioeconomic status and in opioid use. This finding was due to the fact that the prescription of strong opioids tends to be especially notable in the metropolitan area of Barcelona, but it more protocolised and controlled, which may reduce the was also observed in the cities of Girona, Lleida, and Tarragona, differences in consumption between socioeconomic groups. three of the region’s major urban areas. Likewise, in the prescription of opioids for a cancer diagnosis, These socioeconomic differences in opioid use could be socioeconomic status has a much lower impact than in the case of explained by two factors. The first possible explanation a diagnosis of non-oncological pain. There are two possible concerns the differences between the individual socioeconomic explanations for this. First, cancer is a particularly sensitive characteristics of the inhabitants of the different HAs. Previous issue for the healthcare system and its treatment follows literature reports have highlighted the wide-ranging impact of clearly established guidelines, with the result that the influence socioeconomic conditions on the population’s health (García- of socioeconomic factors is probably less important than in other Altés et al., 2018; Kivimäki et al., 2020). In the specific case of situations. Second, the data on drug prescription used in this CNCP, the risk of suffering this type of pain is higher in study corresponded to drugs dispensed at a community unfavourable socioeconomic conditions (Riskowski, 2014; pharmacy office; it was not possible to obtain data on Grol-Prokopczyk, 2017; Jonsdottir et al., 2019), a circumstance consumption in the hospital or long-term care setting. Despite that may increase opioid prescription. The second explanation the fact that socioeconomic status has a lesser impact on the relates to the differences in the characteristics of each HA. Factors prescription of opioids for oncological pain, our results indicated such as the concentration of poverty and social segregation, the some notably high values in the most disadvantaged HAs, physical structure of the environment (e.g., adaptability for especially in the prescription of strong opioids. On this point, people with disabilities, accessibility to public transport, etc.) previous research has pointed out that an individual’s or the network of public facilities may have a decisive influence on socioeconomic status affects their risk of developing cancer the prevalence of morbidities and pain (Marí-Dell’Olmo et al., (Mihor et al., 2020). 2016; Rocha et al., 2017). From a gender perspective, the results show huge differences Other important issues for debate are environmental or in the consumption of opioids between men and women. The individual differences such as the access to supplementary issue of gender inequalities in health has been extensively studied; treatments (such as additional physiotherapy or exercise women have been found to present worse indicators of perceived programmes not covered by public health services) and the health and higher prevalence of certain pathologies (Mauvais- access to pain treatment specialists in different HAs. These Jarvis et al., 2020) and pain (Fillingim et al., 2009; Andrews et al., treatments may be less accessible to some people, either for 2018; Sylwander et al., 2020). Moreover, women tend to make economic reasons, lack of time, or because they cannot greater use of health services (García-Altés et al., 2018; Mihor prioritize their own care. Differences between HAs may also et al., 2020). Regarding opioid consumption, areas with lower be driven by the availability of multidisciplinary services in pain Gender Development Index scores have the worst results in management. In this context, the results of the study suggest that analgesic management indicators. Women are prescribed more opioid analgesics have become a fast and economical way for the opioids than men but make fewer visits to specialist doctors Catalan health system to treat CNCP. In addition, opioid (Chilet-Rosell et al., 2013). Differences in attitudes to opioid management differs according to the approaches of the management between healthcare units and professionals have medical teams and the guidelines in place in each HA, which also been reported (Green et al., 2003; Perelló Bratescu et al., may explain some of the differences in use patterns. Previous 2020). What remains unclear is the relative contribution of studies in the United Kingdom, which like Catalonia has a biological sex to pain mechanisms and its distribution and the universal healthcare system, have identified similar risk factors contribution of gender (Fillingim et al., 2009). A gender in opioid prescribing, including the lack of time spent on perspective approach is needed to accumulate more evidence dedicated patient care and socioeconomic deprivation (Curtis and clarify its social influence to chronic pain and its et al., 2019). In contrast, one study in Mexico showed that areas management (Serdarevic et al., 2017; Mauvais-Jarvis et al., 2020). with a higher socio-economic status had higher rates of opioid In interpreting the results, the greater proportion of women in dispensing after adjusting for palliative care need. Factors such as the population should be borne in mind. This difference becomes the coexistence of different health systems and differences in wider at more advanced ages, as women in the over-75 age group access both to medicines and to health services may be the main outnumber men by 50%. However, data on individual opioid reasons for these disparities, with the most vulnerable areas being users showed that, in the same population, there were more Frontiers in Pharmacology | www.frontiersin.org 8 October 2021 | Volume 12 | Article 750193 Serra-Pujadas et al. Opioids use Inequalities in Catalonia female than male users in absolute and relative terms for both include records of all public healthcare services and strong and weak opioids, thus maintaining the gender differences. prescriptions dispensed for the whole Catalan population. The results of the study show that the socioeconomic gradient These sources provided valuable information and are in opioid use is much more pronounced in women than in men. increasingly used as a robust tool in public health research. Gender should be established as a key social factor in opioid To conclude, our results shed light on the social differences consumption for two reasons: the gender bias in the health system existing between geographical areas in the management of and the poorer health status of women, partly due to a higher opioid use in CNCP. Our findings calls for stronger action to manifestation of pain symptoms and partly due to their less promote best practices in prescribing for chronic pain and to favourable socioeconomic status. These dimensions are reduce socioeconomic and gender variation between intersectional and explain, among other things, the large geographical areas. This study also provides a model for differences observed between men and women (Serdarevic routine monitoring of opioid prescribing in order to et al., 2017; Mauvais-Jarvis et al., 2020). promote interventions able to reduce high-dose This study has some limitations. First of all, as it is an consumption in identified HAs. ecological study, it is impossible to determine the effect that the various factors have on the use of opioid analgesics at the individual level. Also, within an HA with an assigned ISC, there DATA AVAILABILITY STATEMENT might be individuals with very different characteristics. Second, the 1 year time period used in the study does not allow us to The original contributions presented in the study are available through determine trends in opioid use. Third, cancer prevalence and the the Instamaps website. https://www.instamaps.cat/visor.html? patients-per-physician ratio in each HA were not taken into businessid586a401e5496f46278c33dbd8360a150&3Dfalse#, account. Finally, the data corresponds to the drugs dispensed further inquiries can be directed to the corresponding author. by the public health system in pharmacies. Therefore, we do not have data on private consumption (i.e., through private insurance or other sources), or data on hospitals or long-term care AUTHOR CONTRIBUTIONS prescriptions. The data on opioid prescribing should also be interpreted with Conceptualisation: NC and AG. Methodology, software care. A high opioid use does not necessarily reflect inadequate and data curation: SS, CA, JS, JF, and NC. Writing: pain control management; it may represent a satisfactory original draft preparation SS, CA, JS, JF, NC, and AG. response of the health system to the requirements of the Supervision: NC and AG. All authors have approved the population. The aim of this study is to highlight the gap final manuscript. existing between use rates due to different socioeconomic and This study was carried out as part of the university degree final sociodemographic factors, and is certainly not to criticise project of SS, CA, JS, and JF at the Universitat Pompeu Fabra, legitimate medical uses. Barcelona. All these authors contributed equally to the Despite the limitations, this study presents a number of conception and execution of the project and share first strengths. We used administrative medical databases that authorship. Colls, C., Mias, M., and García-Altés, A. (2020). [A Deprivation index to Reform REFERENCES the Financing Model of Primary Care in Catalonia (Spain)]. Gac Sanit 34 (1), 44–50. doi:10.1016/j.gaceta.2018.07.015 Agència de Qualitat i Avaluació Sanitàries de Catalunya (2018). Guia Curtis, H. J., Croker, R., Walker, A. J., Richards, G. C., Quinlan, J., and Goldacre, B. D’atenció Als Pacients Amb Dolor Crònic No Oncològic Utilitzant (2019). Opioid Prescribing Trends and Geographical Variation in England, Analgèsics Opioides. Available at: https://aquas.gencat.cat/web/.content/ 1998-2018: a Retrospective Database Study. Lancet Psychiatry 6 (2), 140–150. minisite/aquas/publicacions/2018/guia_atencio_pacients_dolor_opioides_ doi:10.1016/S2215-0366(18)30471-1 vc_aquas2018ca.pdf. European Monitoring Centre for Drugs and Drug Addiction (2020). European Agencia Española de Medicamentos y Productos Sanitarios (). Utilización de Drug Report 2020: Trends and Developments. Luxembourg: Publications Office medicamentos opioides en España durante el periodo 2010–2018. Available at: of the European Union. doi:10.2810/123451 https://www.aemps.gob.es/medicamentos-de-uso-humano/observatoriode-uso- Fillingim, R. B., King, C. D., Ribeiro-Dasilva, M. C., Rahim-Williams, B., and Riley, de-medicamentos/utilizacion-de-medicamentos-opioides-en-espana-durante-el- J. L. (2009). Sex, Gender, and Pain: A Review of Recent Clinical and periodo-2010-%0A2018/.%0A. Experimental Findings. J. Pain 10 (5), 447–485. doi:10.1016/j.jpain.2008.12.001 Andrews, P., Steultjens, M., and Riskowski, J. (2018). Chronic Widespread Pain Friedman, J., Kim, D., Schneberk, T., Bourgois, P., Shin, M., Celious, A., et al. Prevalence in the General Population: A Systematic Review. Eur. J. Pain 22 (1), (2019). Assessment of Racial/Ethnic and Income Disparities in the Prescription 5–18. doi:10.1002/ejp.1090 of Opioids and Other Controlled Medications in California. JAMA Intern. Med. Bosetti, C., Santucci, C., Radrezza, S., Erthal, J., Berterame, S., and Corli, O. (2019). 179 (4), 469–476. doi:10.1001/jamainternmed.2018.6721 Trends in the Consumption of Opioids for the Treatment of Severe Pain in García-Altés, A., Ruiz-Muñoz, D., Colls, C., Mias, M., and Martín Bassols, N. Europe, 1990-2016. Eur. J. Pain 23 (4), 697–707. doi:10.1002/ejp.1337 (2018). Socioeconomic Inequalities in Health and the Use of Healthcare Services in Chilet-Rosell, E., Ruiz-Cantero, M. T., Sáez, J. F., and Álvarez-Dardet, C. Catalonia: Analysis of the Individual Data of 7.5 Million Residents. J. Epidemiol. (2013). Inequality in Analgesic Prescription in Spain. A Gender Community Health 72 (10), 871–879. doi:10.1136/jech-2018-210817 Development Issue. Gac Sanit 27 (2), 135–142. doi:10.1016/ Goldberg, D. S., and McGee, S. J. (2011). Pain as a Global Public Health Priority. j.gaceta.2012.04.014 BMC Public Health 11, 770. doi:10.1186/1471-2458-11-770 Frontiers in Pharmacology | www.frontiersin.org 9 October 2021 | Volume 12 | Article 750193 Serra-Pujadas et al. Opioids use Inequalities in Catalonia Goodman-Meza, D., Friedman, J., Kalmin, M. M., Aguilar-Posada, E., Seamans, M. Park, J. Y., and Wu, L. T. (2020). Sources of Misused Prescription Opioids and J., Velazquez-Moreno, S., et al. (2021). Geographical and Socioeconomic Disparities Their Association with Prescription Opioid Use Disorder in the United States: in Opioid Access in Mexico, 2015-19: a Retrospective Analysis of Surveillance Data. Sex and Age Differences. Subst. Use Misuse 55 (6), 928–936. doi:10.1080/ Lancet Public Health 6 (2), e88–e96. doi:10.1016/S2468-2667(20)30260-7 10826084.2020.1713818 Green, C. R., Wheeler, J. R., and LaPorte, F. (2003). Clinical Decision Making in Perelló Bratescu, A., Adriyanov, B., Dürsteler, C., Sisó-Almirall, A., Álvarez Pain Management: Contributions of Physician and Patient Characteristics to Carrera, M. A., and Riera Nadal, N. (2020). Opioides fuertes y dolor Variations in Practice. J. Pain 4 (1), 29–39. doi:10.1054/jpai.2003.5 crónico no oncológico en Cataluña. Análisis del patrón de prescripción por Grol-Prokopczyk, H. (2017). Sociodemographic Disparities in Chronic Pain, Based parte de los médicos de familia. Revista Española de Anestesiología y on 12-year Longitudinal Data. Pain 158 (2), 313–322. doi:10.1097/ Reanimación 67 (2), 68–75. doi:10.1016/j.redar.2019.08.003 j.pain.0000000000000762 Riskowski, J. L. (2014). Associations of Socioeconomic Position and Pain Prevalence in HCUP CCS (2017). Healthcare Cost and Utilization Project (HCUP). Rockville, the united states: Findings from the National Health and Nutrition Examination MD: Agency for Healthcare Research and Quality. www.hcup-us.ahrq.gov/ Survey. Pain Med. 15 (9), 1508–1521. doi:10.1111/pme.12528 toolssoftware/ccs/ccs.jsp. Rocha, V., Ribeiro, A. I., Severo, M., Barros, H., and Fraga, S. (2017). Institut Cartografic Generalitat Catalunya (n.d.). Instamaps. Available at: https:// Neighbourhood Socioeconomic Deprivation and Health-Related Quality of www.instamaps.cat/#/ (Retrieved March 19, 2021). Life: A Multilevel Analysis. PLoS ONE 12 (12), e0188736. doi:10.1371/ Jonsdottir, S., Ahmed, H., Tómasson, K., and Carter, B. (2019). Factors Associated journal.pone.0188736 with Chronic and Acute Back Pain in Wales, a Cross-Sectional Study. BMC Santoro, T. N., and Santoro, J. D. (2018). Racial Bias in the US Opioid Epidemic: A Musculoskelet. Disord. 20 (1), 215. doi:10.1186/s12891-019-2477-4 Review of the History of Systemic Bias and Implications for Care. Cureus 10 Kahan, M., Wilson, L., Mailis-Gagnon, A., and Srivastava, A. (2011). Canadian (12), e3733. doi:10.7759/cureus.3733 Guideline for Safe and Effective Use of Opioids for Chronic Noncancer Pain: Schieber, L. Z., Guy, G. P., Seth, P., Young, R., Mattson, C. L., Mikosz, C. A., et al. Clinical Summary for Family Physicians. Part 2: Special Populations. Can. Fam. (2019). Trends and Patterns of Geographic Variation in Opioid Prescribing Physician 57 (11), 1269–e428. Practices by State, United States, 2006-2017. JAMA Netw. Open 2 (3), e190665. Kalkman, G. A., Kramers, C., van Dongen, R. T., van den Brink, W., and doi:10.1001/jamanetworkopen.2019.0665 Schellekens, A. (2019). Trends in Use and Misuse of Opioids in the Scholl, L., Seth, P., Kariisa, M., Wilson, N., and Baldwin, G. (2018). Drug and Netherlands: a Retrospective, Multi-Source Database Study. Lancet Public Opioid-Involved Overdose Deaths - United States, 2013-2017. MMWR Morb Health 4 (10), e498–e505. doi:10.1016/S2468-2667(19)30128-8 Mortal Wkly Rep. 67 (11), 1419–1427. doi:10.15585/mmwr.mm675152e1 Kivimäki, M., Batty, G. D., Pentti, J., Shipley, M. J., Sipilä, P. N., Nyberg, S. T., et al. Scottish Intercollegiate Guidelines Network (SIGN) (2013). “Management of (2020). Association between Socioeconomic Status and the Development of Chronic Pain,”. A National Clinical Guideline (Issue December). Mental and Physical Health Conditions in Adulthood: a Multi-Cohort Study. Serdarevic, M., Striley, C. W., and Cottler, L. B. (2017). Sex Differences in Lancet Public Health 5 (3), e140–e149. doi:10.1016/S2468-2667(19)30248-8 Prescription Opioid Use. Curr. Opin. Psychiatry 30 (4), 238–246. Labianca, R., Sarzi-Puttini, P., Zuccaro, S. M., Cherubino, P., Vellucci, R., and doi:10.1097/YCO.0000000000000337 Fornasari, D. (2012). Adverse Effects Associated with Non-opioid and Opioid Song, Z. (2017). Mortality Quadrupled Among Opioid-Driven Hospitalizations, Treatment in Patients with Chronic Pain. Clin. Drug Investig. 32 Suppl 1 (Suppl. Notably within Lower-Income and Disabled white Populations. Health Aff. 1), 53–63. doi:10.2165/11630080-000000000-00000 (Millwood) 36 (12), 2054–2061. doi:10.1377/hlthaff.2017.0689 Lande, A. (1962). The Single Convention on Narcotic Drugs, 16. New York, NY: StataCorp (n.d.). Stata: Software for Statistics and Data Science. International Organization, 776–797. doi:10.1017/S0020818300011620The Sylwander, C., Larsson, I., Andersson, M., and Bergman, S. (2020). The Impact of Single Convention on Narcotic Drugs, 1961Int. Org. Chronic Widespread Pain on Health Status and Long-Term Health Predictors: Majors-Foley, K. (2016). Use of and Barriers to Access to Opioid Analgesics: A A General Population Cohort Study. BMC Musculoskelet. Disord. 21 (1), 36. Worldwide, Regional, and National Study. J. Emerg. Med. 51 (1), 93. doi:10.1186/s12891-020-3039-5 doi:10.1016/j.jemermed.2016.06.023 U.S. Department of Health and Human Services (n.d.). What Is the U.S. Opioid Manchikanti, L., Abdi, S., Atluri, S., Balog, C. C., Benyamin, R. M., Boswell, M. V., Epidemic? Available at: https://www.hhs.gov/opioids/about-the-epidemic/ et al. (2012). American Society of Interventional Pain Physicians (ASIPP) index.html (Retrieved March19, 2021). Guidelines for Responsible Opioid Prescribing in Chronic Non-cancer Pain: WHO Collaborating Centre for Drug Statistics (n.d.). ATC/DDD Index. Available Part 2--guidance. Pain Physician 15 (3 Suppl. l), S67–S116. at: https://www.whocc.no/atc_ddd_index/ (Retrieved March 19, 2021). Manchikanti, L., Boswell, M. V., and Hirsch, J. A. (2013). Lessons Learned in the World Health Organisation (1996). The Journal of Infectious Diseases. 2nd ed. Abuse of Pain-Relief Medication: A Focus on Healthcare Costs. Expert Rev. Available at: https://apps.who.int/iris/handle/10665/37896 (Retrieved March Neurother 13 (5), 527–544. doi:10.1586/ern.13.33 19, 2021). Cancer Pain Relief: A Guide to Opioid Availability. Marí-Dell’Olmo, M., Gotsens, M., Palència, L., Rodríguez-Sanz, M., Martinez- Beneito, M. A., Ballesta, M., et al. (2016). Trends in Socioeconomic Inequalities Conflict of Interest: The authors declare that the research was conducted in the in Mortality in Small Areas of 33 Spanish Cities. BMC Public Health 16 (1), 663. absence of any commercial or financial relationships that could be construed as a doi:10.1186/s12889-016-3190-y potential conflict of interest. Mauvais-Jarvis, F., Bairey Merz, N., Barnes, P. J., Brinton, R. D., Carrero, J. J., DeMeo, D. L., et al. (2020). Sex and Gender: Modifiers of Health, Disease, and Publisher’s Note: All claims expressed in this article are solely those of the authors Medicine. Lancet 396 (10250), 565–582. doi:10.1016/S0140-6736(20)31561-0 and do not necessarily represent those of their affiliated organizations, or those of Mihor, A., Tomsic, S., Zagar, T., Lokar, K., and Zadnik, V. (2020). Socioeconomic the publisher, the editors and the reviewers. Any product that may be evaluated in Inequalities in Cancer Incidence in Europe: A Comprehensive Review of this article, or claim that may be made by its manufacturer, is not guaranteed or Population-Based Epidemiological Studies. Radiol. Oncol. 54 (1), 1–13. endorsed by the publisher. doi:10.2478/raon-2020-0008 Morden, N. E., Munson, J. C., Colla, C. H., Skinner, J. S., Bynum, J. P., Zhou, W., Copyright © 2021 Serra-Pujadas, Alonso-Buxadé, Serra-Colomer, Folguera, et al. (2014). Prescription Opioid Use Among Disabled Medicare Beneficiaries: Carrilero and García-Altés. This is an open-access article distributed under the Intensity, Trends, and Regional Variation. Med. Care 52 (9), 852–859. terms of the Creative Commons Attribution License (CC BY). The use, distribution doi:10.1097/MLR.0000000000000183 or reproduction in other forums is permitted, provided the original author(s) and the Pando, T. de., Molina, A., Carbonell, P., and Delgadillo, J. (2017). Informe copyright owner(s) are credited and that the original publication in this journal is d’utilització de medicaments opioides a Catalunya: anàlisi del període 2012- cited, in accordance with accepted academic practice. No use, distribution or 2016. Scientia 1–11. reproduction is permitted which does not comply with these terms. Frontiers in Pharmacology | www.frontiersin.org 10 October 2021 | Volume 12 | Article 750193