Research Article Eur Addict Res Received: June 25, 2021 Accepted: December 3, 2021 DOI: 10.1159/000521425 Published online: January 17, 2022 COVID-19 Lockdown and Consumption Patterns among Substance Use Disorder Outpatients: A Multicentre Study Lara Grau-López a, b, c, d Constanza Daigre a, b, c, d Raúl Felipe Palma-Alvarez a, b, c, d Marta Sorribes-Puertas a, b, c, d Pedro Serrano-Pérez a, b, c, d Marta Quesada-Franco a, b, c, d Lidia Segura e Marta Coronado e Josep Antoni Ramos-Quiroga a, b, c, d Joan Colom e XAD-COVID19-Group e aDepartment of Psychiatry, Addiction and Dual Diagnosis Section, Hospital Universitari Vall d’Hebron, Barcelona, Spain; bPsychiatry Group, Mental Health and Addiction, Vall d’Hebron Research Institute (VHIR), Barcelona, Spain; cBiomedical Network Research Centre on Mental Health (CIBERSAM), Madrid, Spain; dDepartment of Psychiatry and Forensic Medicine, Autonomous University of Barcelona, Barcelona, Spain; eSubdirecció general de Drogodependències, Agència de Salut Pública de Catalunya, Barcelona, Spain Keywords tern). Results: More than 62% of the patients maintained or Anxiety · Consumption patterns · COVID-19 · Dual diagnosis · worsened their consumption pattern during confinement, Lockdown · Pandemic · Substance use disorder and about 38% improved throughout this time. Diverse fac- tors were associated with the changes in pattern like age, addiction severity, psychological distress during lockdown, Abstract social and familial issues, and therapeutic variables. Conclu- Introduction: COVID-19 has had a great impact on mental sion: The home lockdown during the COVID-19 pandemic health in the general population. However, few studies have was associated with major implications for substance con- focused on patients with substance use disorders (SUDs). sumption and psychiatric distress among SUD outpatients. This research aimed to compare the clinical status and sub- Considering this, the need to plan appropriate interventions stance use patterns of SUD outpatients, before and during in cases of similar health crises is highlighted. confinement due to the COVID-19 pandemic. Methods: This © 2022 The Author(s). multicentre study recruited 588 patients (29.2% women; M Published by S. Karger AG, Basel age = 47.4 ± 11.7 years) in thirteen centres for addiction treatment in Catalonia, Spain. All subjects were evaluated with validated instruments and ad hoc questionnaires. The Lara Grau-López and Constanza Daigre contributed equally to this sample was divided into 3 groups according to how the sub- manuscript. stance consumption pattern changed during lockdown Collaborative Addiction Research Group of the Agència de Salut (maintained pattern, worsened pattern, and improved pat- Pública de Catalunya (full list in the Appendix). karger@karger.com © 2022 The Author(s). Correspondence to: www.karger.com/ear Published by S. Karger AG, Basel Raúl Felipe Palma-Alvarez, rpalma @ vhebron.net This article is licensed under the Creative Commons Attribution 4.0 International License (CC BY) (http://www.karger.com/Services/ OpenAccessLicense). Usage, derivative works and distribution are permitted provided that proper credit is given to the author and the original publisher. Introduction use of legal substances. One study, carried out with alco- hol-dependent patients, found an increase in alcohol con- The COVID-19 pandemic has affected the entire world sumption [20]. Another study, using hair analysis, con- [1]. Spain and also Catalonia were among the most affected ducted on 30 subjects, showed a decrease in the consump- areas in the first months of the pandemic, which has caused tion of illegal drugs but a significant increase in an unprecedented crisis with significant health, social, and consumption of benzodiazepines and alcohol, substances economic repercussion [2]. To halt the expansion of CO- which are more easily accessible [21]. These data were VID-19 in Spain, a nationwide at-home lockdown was re- confirmed in a separate study from Spain which found an quired for the general population, which forced a rapid ad- increase in alcohol and benzodiazepine consumption in aptation of health measures to guarantee that essential ac- drug-dependent individuals [5]. Few studies include il- tivities could continue to take place, including implementing legal substance use among patients with SUD. In the telematic assistance, among other measures [3–6]. USA, a significant increase in illicit drug use (cocaine, This exceptional situation has been related to an in- fentanyl, heroin, and methamphetamine) was observed crease in anxiety and depression in the general public [7– through positive urine drug tests in a population diag- 9], as well as in more vulnerable communities, such as nosed with, or at risk for, SUD during the 4 months after patients with previous mental disorders [10, 11]. Patients the COVID-19 pandemic declaration [22]. An increase in with substance use disorders (SUD) often face other psy- overdoses has also been described in the emergency de- chiatric pathologies at the same time, which has impor- partments in the USA during the pandemic [23]. How- tant connotations for the aetiology, evolution, and treat- ever, despite these studies, there are little data on the im- ment of both disorders [12]. This co-occurrence of disor- pact of lockdown on patients being treated for SUD, es- ders is known as dual diagnosis, which emphasizes the pecially those who consume illegal drugs. interrelatedness of different disorders [13]. Dual diagno- In terms of COVID-19 infection, SUD patients were sis patients are part of a heterogeneous and complex considered an at-risk group [24]. Diverse publications group, with high rates of relapse and more severe symp- have warned about greater contagion risk and physical toms of addiction [14]. consequences among those with ill health and a poor so- Regarding substance use in the general population cial situation [25–27]. Nonetheless, similar incidence of during lockdown, changes in drinking behaviours have severe COVID-19 effects has been reported for SUD pa- been reported [15]. Two large Web-based surveys found tients and for people without SUD [5, 26, 28], although that alcohol and tobacco use increased in this population studies are needed to analyse the risk of COVID-19 infec- but not use of other substances [16, 17]. One study re- tion in SUD population. ported that individuals consumed slightly more alcohol This study aimed to compare the clinical status and and also smoked marginally more cigarettes compared to substance use patterns, before and during confinement the period before the lockdown [16]. Another study, that due to the COVID-19 pandemic, in SUD patients treated analysed changes in alcohol drinking behaviour and to- in outpatient centres for drug addiction in Catalonia. An bacco smoking in the general population, found that analysis was also done of clinical- and COVID-19-related 35.5% of the study population reported an increase in variables in patients that improved, maintained, or wors- drinking during the lockdown [17]. ened their consumption pattern during confinement Since the beginning of the pandemic, hypotheses have compared to their status before. been proposed on how lockdown could affect patients be- ing treated for addictions. However, it was unknown how this historic health and social milestone would affect this Materials and Methods vulnerable population. On one hand, it was proposed Study Design and Patients that, as had occurred in previous pandemics, substance A multicentre, descriptive study was conducted on SUD pa- use would increase [18] or by not being able to access tients who received treatment in outpatient centres for drug addic- usual treatments [19]. On the other hand, it was hypoth- tion in Catalonia. Thirteen centres participated in the study, and esized that the decreased availability of substances may they collected retrospective information about clinical features lessen their consumption [18]. and substance use information during the 3-month lockdown, from March 15, 2020, to May 24, 2020. Recruitment lasted from Most published studies to date suggest there was an March 2020 to September 2020. The inclusion criteria were people increase in substance use among SUD patients during the with a diagnosis of SUD aged older than 18 years. Patients with low first months of the COVID-19 pandemic, especially in the Spanish proficiency that interfered with their ability to understand 2 Eur Addict Res Grau-López et al. DOI: 10.1159/000521425 the study proposal were excluded. The project was approved by the Table 1. Sociodemographic and clinical variables prior to lockdown Ethics Committee of Vall d’Hebron Hospital (PR-(AG)386–2020). in Catalonia in the total sample (n = 588) The legal department of the Sub-Directorate General of Drug De- pendence of Health/Public Health Agency of Catalonia approved Prior to lockdown the database. Patients did not receive any financial compensation, and written informed consent was obtained from all participants. Sociodemographic characteristics, n (%) In the case of telephone visits, consent was procured orally and Gender (men) 416 (70.7) ratified in writing in the following in-person visit. Mean age, years 47.4±11.7 Spanish nationality 526 (89.5) Procedure Secondary or higher education 278 (50.5) After obtaining written informed consent from the partici- Marital status (married or partner) 363 (65.9) pants, the psychiatrists or psychologists in charge of each partici- Living with their family 399 (70.2) pant interviewed them using an ad hoc questionnaire to evaluate Employed 149 (26.2) their clinical and substance use status both before and during the Criminal records 179 (30.4) lockdown caused by the COVID-19 pandemic. The interviews History of SUD, n (%) were devised through either in-person visits or telematically dur- Primary diagnosis ing the patient’s usual programmed follow-up visits. All the infor- OUD 145 (24.7) mation was entered into a database following the anonymization CUD 243 (41.3) of patients and professionals. Vall d’Hebron Hospital and the Sub- AUD 333 (56.6) Directorate General of Drug Dependence of Health/Public Health BUD 57 (9.7) Agency of Catalonia coordinated the study and the data collection. CNUD 103 (17.5) TUD 87 (14.8) Instruments and Variables Poly-SUDs 204 (32.7) Previous Sociodemographic and Clinical Features Injecting drug use 94 (16) An ad hoc designed interview was used to evaluate demograph- Previous treatment for SUD 455 (74.7) ic and clinical data at the time of enrolment. Sociodemographic Abstinence for more than a year 188 (32) features recorded were gender, age, nationality, civil status, hous- Dual diagnosis, n (%) ing, educational level, employment status, and criminal record. In- Dual diagnosis 394 (67) formation regarding SUD included primary diagnosis, whether Psychotic disorder 74 (12.6) there was polysubstance use (understood as 3 or more SUDs), drug Depressive disorder 219 (37.2) administration route, previous SUD treatments, and length of ab- Anxiety disorder 129 (21.9) stinence prior to lockdown and the time of abstinence at the begin- ADHD 26 (4.4) ning of lockdown. Dual diagnosis was assessed by a trained psy- Personality disorders 119 (20.2) chiatrist or clinical psychologist and established by clinical judge- Any medical condition 378 (64.3) ment, following the DSM-5 criteria. After dual diagnosis assessment, the mental disorders diagnosed were grouped by psy- ADHD, attention deficit hyperactivity disorder; AUD, alcohol use chotic, depressive, anxiety, and personality disorders. disorder; CNUD, cannabis use disorder; CUD, cocaine use disorder; OUD, opiate use disorder; SUD, substance use disorder; TUD, tobac- Substance Use Status, COVID-19-Related Variables, and co use disorder. Psychopathological Variables during Lockdown Substance use pattern during the lockdown was assessed by the same ad hoc designed questionnaire. This questionnaire included the evaluation of SUDs before and during the lockdown, history of SUD, changes in substance consumption during lockdown, clini- confirmed COVID-19 infection, hospitalization due to CO- cal variables regarding psychiatric status, and COVID-19-related VID-19, breaking lockdown norms, working during lockdown, variables (described in Tables 1, 2). Considering changes in sub- decreased income because of the pandemic, changes in housing stance consumption (in terms of the amount consumed, relapses status, family relationships, social matters, and health status. Also or abstinence) during the lockdown (compared to before), par- recorded was whether the treatment was telematic or in person and ticipants were classified into 3 groups: maintained pattern (pa- how patients valued these treatments. tients whose consumption remained the same in terms of amount, Finally, psychopathological variables during lockdown were frequency, and type of consumed substance), worsened pattern measured, such as insomnia symptoms, eating disorder symp- (patients who relapsed in their use of the main or another sub- toms, sexual alterations, and attendance at the psychiatric emer- stance or increased consumption, with the exception of tobacco), gency room for SUD or other matters. Moreover, the following and improved pattern (patients who achieved or maintained absti- instruments were used to assess anxiety symptoms, general psychi- nence or those who consumed less frequently or in lower amounts atric status, and clinical global impression during lockdown: compared to previous substance consumption). The types of sub- • Overall psychiatric severity was evaluated through the Clinical stances consumed during the lockdown and craving levels were Global Impression-Severity scale (CGI-S [29]) and Clinical also recorded. Global Impression-Improvement scale (CGI-I [29]) which as- COVID-19-related variables were measured, including vari- sess improvement and worsening during lockdown. They im- ables such as concern due to COVID-19, COVID-19 symptoms, plement the widely used Likert scale (0–7 points). COVID-19 Lockdown and Consumption Eur Addict Res 3 Patterns DOI: 10.1159/000521425 Table 2. Clinical and COVID-19-related variables during lockdown • The Clinical Anxiety Scale (CAS) [30] is a hetero-applied scale in Catalonia in the total sample (n = 588) which measures anxiety symptoms during lockdown using 7 items that are scored on a 5-point Likert scale (0–4). Five or During lockdown more points indicate the presence of anxiety symptoms. It has adequate validity, reliability, and sensitivity to change. Assess- SUD variables, n (%) ment of anxiety symptoms was included because anxiety symp- Consumption pattern evolution toms can be induced by the COVID-19 crisis and because they Active consumption maintained 148 (25.2) are frequently present in people with addiction. Worsened consumption pattern 217 (36.9) • The Brief Psychiatric Rating Scale (BPRS) [31] assesses 18 Improved consumption or abstinence 223 (37.9) symptom domains through clinical judgement and question- Substance use increases, n (%) ing, a Likert scale is used from 0 (not present) to 6 (extremely Opiate use increases in OUD 18 (12.4) severe). The evaluated domains are somatic concern, anxiety, Cocaine use increases in CUD 60 (24.7) emotional withdrawal, conceptual disorganization, guilt feel- Alcohol use increases in AUD 80 (24) ings, tension, mannerisms and posturing, grandiosity, depres- Benzodiazepine use increases in BUD 18 (31.6) sive mood, hostility, suspiciousness, hallucinatory behaviour, Cannabis use increases in CNUD 30 (29.1) motor retardation, uncooperativeness, unusual thought con- Tobacco use increases in TUD 24 (27.6) tent, blunted affect, excitement, and disorientation. The evalu- Substance used during lockdown, n (%) ation takes 10 min, and following the interview with the pa- Opiate consumption 34 (5.8) tient, the clinician rates each item on the Likert scale through Cocaine consumption 100 (17) observation and questioning depending on the assessed item. Alcohol consumption 176 (29.9) Cannabis consumption 114 (19.4) Statistical Analysis Benzodiazepine consumption 63 (10.7) Descriptive statistics (mean, standard deviation, and frequency Hospital emergency for SUD 26 (4.4) tables) of the main variables were calculated. The data were then Withdrawal symptoms 73 (12.4) analysed at the bivariate level in order to compare the 3 groups es- Increased craving 213 (36.2) tablished according to the consumption pattern during lockdown. COVID-19-related variables, n (%) The χ2 test was used to compare categorical variables and the COVID-19 concern 342 (58.2) ANOVA test for continuous variables when 3 groups were com- COVID-19 symptoms 77 (13.1) pared. The χ2 test was considered not applicable when one or more Positive COVID-19 8 (1.4) of the cells had an expected count <5. To reduce false-positive re- Hospitalization because of COVID-19 2 (0.3) sults, the Bonferroni correction for multiple tests was performed Broke the lockdown norm 206 (35) according to the number of tests in each bivariate analysis and to Working 100 (17) avoid type 1 error. Decreased income 170 (28.9) Only variables considered clinically relevant, those which re- Housing changed 49 (8.3) tained statistical significance after Bonferroni correction and with- Worsened family relationship 139 (23.7) out multicollinearity, were included in the multinomial regression Unresolved social matters 80 (13.8) models. The compared groups by the dependent variable were im- Self-reported impaired health 145 (24.7) proved pattern, maintenance pattern, and worsened pattern. At Telematic treatment 237 (40.3) the bivariate level, men and women were analysed together be- Positively valued treatment 426 (72.4) cause there were no significant differences between genders in the Psychiatric status, n (%) dependent variable; however, in the multivariate analysis, gender Insomnia symptoms 260 (44.2) and age were included in the models performed. All statistical hy- Eating disorder symptoms 286 (48.7) potheses were two-sided, and a p value of <0.05 was considered Sexual behaviour alterations 85 (14.5) statistically significant. SPSS version 20 (SPSS Inc., Armonk, NY, Psychiatric emergencies 29 (4.9) USA) for Windows was used for all analyses. Anxiety symptoms (CAS) 333 (60) Panic attacks 70 (11.9) BPRS (above average 27) 254 (43.2) Depressive symptoms in BPRS 208 (35.4) CGI-S (greater severity 5–7) 69 (11.8) Results Worsening global impression of change 237 (40.4) Sample Recruitment and Sample Features AUD, alcohol use disorder; BPRS, Brief Psychiatric Rating Scale; A total of 588 patients were enrolled, representing BUD, benzodiazepine use disorder; CGI-S, Clinical Global Impres- 9.1% of the patients treated in the 13 participating cen- sion-Severity scale; CNUD, cannabis use disorder; CUD, cocaine use disorder; OUD, opiate use disorder; SUD, substance use disorder; tres. During the recruitment period, the patients who at- TUD, tobacco use disorder. tended or were visited telematically were consecutively invited to participate. It is likely that due to restrictions derived from the pandemic, 38.4% were not visited dur- ing the recruitment period. Furthermore, the clinicians 4 Eur Addict Res Grau-López et al. DOI: 10.1159/000521425 Table 3. Consumption pattern during, compared to before, lockdown by sociodemographic characteristics, SUDs, and dual diagnosis Improved Maintained Worsened Statistic p value pattern pattern pattern (n = 223) (n = 148) (n = 217) Sociodemographic characteristics, n (%) Gender (men) 154 (69.1) 104 (70.3) 158 (72.8) 0.8 0.68 Mean age, years 49.6±12.2 48.7±11.6 44.3±10.6 13.2 0.0001* Spanish nationality 199 (89.2) 135 (91.2) 192 (88.5) 0.7 0.69 Secondary or higher education 119 (54.8) 63 (44.7) 96 (49.7) 3.6 0.166 Marital status (married or partner) 146 (67.3) 102 (72.3) 115 (59.6) 6.2 0.045 Living with their family 155 (71.4) 110 (75.9) 134 (60) 5.0 0.082 Employed 51 (23.5) 41 (28.3) 57 (27.7) 1.4 0.504 Criminal record 55 (24.7) 45 (30.4) 79 (36.4) 7.2 0.028 SUDs, n (%) Primary diagnosis OUD 39 (17.5) 41 (27.7) 65 (30) 10.2 0.006 CUD 65 (29.1) 50 (33.8) 128 (59) 45.0 0.0001* AUD 135 (60.5) 79 (53.4) 119 (54.8) 2.3 0.315 BUDs 11 (4.9) 14 (9.5) 32 (14.7) 12.1 0.002 CNUD 28 (12.6) 29 (19.6) 46 (21.2) 6.3 0.043 TUD 34 (15.2) 15 (10.1) 37 (17.5) 3.9 0.145 Poly-SUDs 44 (19.7) 56 (37.8) 104 (47.9) 39.5 0.0001* Injecting drug use 20 (9) 24 (16.2) 50 (23) 16.2 0.0001* Previous treatment for SUD 171 (76.7) 114 (77) 170 (78.3) 0.2 0.911 Abstinence for more than a year 93 (41.7) 61 (41.2) 34 (15.7) 42.1 0.0001* Dual diagnosis, n (%) Dual diagnosis 155 (69.5) 91 (61.5) 148 (68.2) 2.8 0.245 Psychotic disorder 27 (12.1) 19 (12.8) 28 (12.9) 0.1 0.963 Depressive disorder 92 (41.3) 49 (33.1) 78 (35.9) 2.8 0.25 Anxiety disorder 51 (22.9) 22 (14.9) 56 (25.8) 6.3 0.042 ADHD 9 (4) 6 (4.1) 11 (5.1) 0.3 0.843 Personality disorders 33 (14.8) 21 (14.2) 65 (30) 20.1 0.0001* Any medical condition 151 (67.7) 95 (64.2) 132 (60.8) 2.3 0.321 ADHD, attention deficit hyperactivity disorder; AUD, alcohol use disorder; BUD, benzodiazepine use disorder; CNUD, cannabis use disorder; CUD, cocaine use disorder; OUD, opiate use disorder; SUD, substance use disorder; TUD, tobacco use disorder. * Significant after Bonferroni correction. did not invite 16.2% of the patients to participate due to stance users. It was found that 67% of the participants had different practical reasons (e.g., limitation of time and in- a dual diagnosis, with the most frequently co-occurring volvement of professionals in supporting COVID-19 pa- disorders being depressive disorders followed by anxiety tients). Finally, 26.8% declined to participate in the study, disorders, and 20.2% had personality disorders (See Ta- and 9.5% were excluded due to a language barrier. ble 1). Table 2 describes variables related to SUD, CO- Each centre provided data on the patients in terms of VID-19, and psychiatric status during lockdown; 25.2% of age, sex, and substances consumed. The sociodemograph- patients were classified in the maintained pattern, 36.9% ic, clinical, and COVID-19-related variables both before in the worsened pattern, and 223 (37.9%) in the improved and during lockdown in Catalonia are shown in Tables 1 pattern group (157 of them maintained previous absti- and 2. Women made up 29.2% of the sample, while 70.7% nence, 24 patients achieved abstinence, and 42 patients were men, and the mean age was 47.4 ± 11.7 years. Regard- reduced substance use during lockdown). It was found ing primary SUD diagnosis before lockdown, it was ob- that 31% of the patients with benzodiazepine use disorder served that alcohol and cocaine use were the most preva- (BUD) reported increased consumption of this substance, lent form of SUD, and 32.7% of the patients were polysub- in the case of cannabis 29% reported increased use, and COVID-19 Lockdown and Consumption Eur Addict Res 5 Patterns DOI: 10.1159/000521425 Table 4. Consumption pattern during, compared to before, lockdown by SUD variables, COVID-19-related variables, and psychiatric status during lockdown Improved Maintained Worsened Statistic p value pattern pattern pattern (n = 223) (n = 148) (n = 217) SUD variables, n (%) Hospital emergency for SUD 5 (2.2) 1 (0.7) 20 (9.2) 19.2 0.0001* Withdrawal symptoms 14 (6.3) 6 (4.1) 53 (24.4) 46.0 0.0001* Opiate consumption 2 (0.9) 4 (2.7) 28 (12.9) 32.5 0.0001* Cocaine consumption 8 (3.6) 10 (6.8) 82 (37.8) 105.9 0.0001* Alcohol consumption 28 (12.6) 23 (15.5) 125 (57.6) 125.9 0.0001* Cannabis consumption 17 (7.6) 20 (13.5) 77 (35.5) 59.0 0.0001* Benzodiazepine consumption 8 (3.6) 8 (5.4) 47 (21.7) 43.4 0.0001* COVID-19-related variables, n (%) COVID-19 concern 132 (59.2) 87 (58.8) 123 (56.7) 0.3 0.854 COVID-19 symptoms 29 (13) 10 (6.8) 38 (17.5) 8.9 0.011 Positive for COVID-19 1 (0.4) 2 (1.4) 5 (2.3) NA NA Broke the lockdown norm 54 (24.2) 44 (29.7) 108 (49.8) 34.0 0.0001* Working 38 (17) 29 (19.6) 33 (15.2) 1.2 0.549 Reduced income 56 (25.1%) 43 (29.1%) 71 (32.7%) 3.1 0.212 Change of address 21 (9.4) 8 (5.4) 20 (9.3) 2.2 0.326 Worsened family relationship 39 (17.5) 26 (17.6) 74 (34.3) 21.2 0.0001* Unresolved social matters 21 (9.5) 11 (7.5) 48 (22.4) 21.7 0.0001* Telematic treatment 92 (41.3) 78 (52.7) 67 (30.9) 17.6 0.0001* Self-reported impaired health 51 (22.9) 24 (16.2) 70 (32.3) 12.8 0.002 Positively valued treatment 176 (78.9) 113 (76.4) 137 (63.1) 15.2 0.0001* Psychiatric status, n (%) Insomnia symptoms 91 (40.8) 50 (33.8) 119 (54.8) 17.5 0.0001* Eating disorders symptoms 101 (45.5) 60 (40.5) 125 (57.6) 11.7 0.003 Alterations of sexual behaviour 29 (13) 19 (12.8) 37 (17.1) 1.9 0.392 Psychiatry emergencies 9 (4) 3 (2) 17 (7.8) 6.9 0.031 Anxiety symptoms (CAS) 105 (49.8) 65 (47.1) 163 (79.1) 50.2 0.0001* Panic attack 23 (10.3) 9 (6.1) 38 (17.5) 11.8 0.003 BPRS (above average 27) 75 (33.6) 49 (33.1) 130 (59.9) 39.2 0.0001* Depressive symptoms in BPRS 64 (28.8) 38 (25.7) 106 (48.8) 27.5 0.0001* CGI-S (grater severity, score 5–7) 16 (7.2) 8 (5.4) 45 (20.7) 27.1 0.0001* Worsening (global impression of change) 57 (25.7) 40 (27) 140 (64.5) 83.4 0.0001* BPRS, Brief Psychiatric Rating Scale; SUD, substance use disorder. * Significant after Bonferroni correction less frequently (12.4%) reported was increased opioid use Comparison of the Consumption Pattern before and in patients with opioid use disorder. For the entire sample, during Lockdown alcohol was the most common substance used, followed Tables 3 and 4 show the comparison of the different by cannabis and cocaine. Regarding the COVID-19-relat- variables studied among the 3 groups. It was found that ed variables, 58.2% of patients reported feeling concerned the patients in the worsened-pattern group were signifi- about the disease, 13.1% reported COVID-19 symptoms, cantly younger, were mainly in treatment for cocaine use and only 1.4% (n = 8) were infected. Around 35% of the disorder, were polysubstance users, used the injected sample acknowledged breaking the lockdown norms. Re- route more frequently, fewer had been abstinent for a garding psychiatric status, 60% reported anxiety symp- year, and had more frequent personality disorders (See toms, 35.4% depressive symptoms, and 40.4% reported Table 3). Table 4 shows that patients whose consumption that their psychiatric symptoms worsened during lock- pattern worsened were more commonly seen for emer- down (See Table 2). gencies due to SUD, faced more withdrawal symptoms, and consumed substances more often. 6 Eur Addict Res Grau-López et al. DOI: 10.1159/000521425 Table 5. Multivariate analysis: previous consumption pattern, COVID-19-related variables, SUD variables, and psychiatric status during lockdown Improved pattern Maintained pattern Improved pattern versus versus versus worsened pattern worsened pattern maintained pattern Wald sig. OR 95% CI Wald sig. OR 95% CI Wald sig. OR 95% CI Model 1: prior to lockdown Gender Age (under/above average 47 years) 7.8 0.005 1.8 1.2–2.7 Poly-SUDs 10.1 0.001 0.4 0.3–0.7 10.3 0.001 0.4 0.2–0.7 Injecting drug use Current CUD 10.6 0.001 0.5 0.3–0.7 15.4 <0.0001 2.7 1.6–4.4 During lockdown Model 2: COVID-19-related variables during lockdown Gender Age (under/above average 47 years) 7.2 0.007 0.6 0.4–0.9 Telematic treatment 7.3 0.007 0.5 0.3–0.8 5.1 0.024 0.6 0.4–0.9 Positively valued treatment 7.5 0.006 0.5 0.3–0.8 Broke the lockdown norm 20.1 <0.001 2.7 1.7–4.1 9.4 0.002 2.1 1.3–3.3 Worsened family relationship 4.6 0.032 1.7 1.0–2.7 4.5 0.035 1.8 1.0–3.1 Unresolved social matters 7.7 0.005 2.3 1.3–4.2 7.6 0.006 2.7 1.3–5.6 Model 3: psychiatric status during lockdown Gender Age (under/above average 47 years) 12.9 <0.001 2.1 1.4–3.2 4.1 0.042 1.6 1–2.4 Insomnia symptoms 4.1 0.04 1.6 1–2.6 Anxiety symptoms (CAS) 14.2 <0.001 0.4 0.2–0.6 13.3 <0.0001 2.9 1.6–5 BPRS (above average 27) 5.3 0.021 0.6 0.4–0.9 Regarding COVID-19-related variables, patients Multinomial Regression Models on Previous whose consumption worsened reported breaking lock- Consumption Pattern, COVID-19-Related Variables, down restrictions more frequently. The percentage of in- SUD Variables, and Psychiatric Status during Lockdown fection was higher among patients whose consumption Table 5 shows multivariate analysis allowing a pair- patterns worsened during lockdown; however, due to the wise comparison of the groups. Model 1 analysed previ- low prevalence, it is not possible to make reliable statisti- ous lockdown variables according to the consumption cal comparisons. In addition, those in this group had pattern (χ2 = 74.6, p < 0.0001, Negelkerke = 0.134). The more family conflicts and unresolved social issues. Pa- variables that retained statistical significance in model 1 tients in this group were treated more frequently in face- were age, poly-SUDs, and current cocaine use disorder. to-face meetings than those in the other groups, who had Model 2 shows independently related variables with more telematic visits. Also, patients whose consumption COVID-19 variables during lockdown (χ2 = 278.1, p < worsened valued the treatment offered less positively. 0.0001, Nagelkerke = 0.445). All variables included in Concerning psychiatric status, it was observed that pa- model 2, except gender, retained statistical significance tients whose consumption pattern worsened experi- (age, telematic treatment, positively valued treatment, enced insomnia and anxiety symptoms more frequently break the lockdown norm, worsened family relationship, and had higher scores in the BPRS, as well as greater se- unresolved social matters). verity in the CGI-S and worsened in terms of CGI-I (See Model 3 analyses the variables related to the patients’ Table 4). psychiatric status during lockdown (χ2 = 90.9, p < 0.0001, Nagelkerke = 0.164). The variables that retained statisti- cal significance in model 3 were age, insomnia symptoms, anxiety symptoms, and BPRS score. COVID-19 Lockdown and Consumption Eur Addict Res 7 Patterns DOI: 10.1159/000521425 Discussion which consumption increased the most compared to pri- or to the pandemic, among patients who previously had This study shows that lockdown in Catalonia during an addiction to these substances. A recent study reported the first phase of the COVID-19 pandemic was associated that cannabis users who increased cannabis and tend to with changes in the consumption pattern of a great part use cannabis to cope are at increased risk of experiencing of SUD patients. Patients who worsened their consump- negative cannabis-related consequences during the CO- tion presented greater social difficulties and psychiatric VID-19 pandemic [37]. The patients with opioid use dis- disturbances. order were the ones who reported increased use less fre- Regarding coronavirus infection, 8 patients (1.4%) re- quently. Around a third of patients with BUD increased ported confirmed virus infection. A prior study per- the consumption of benzodiazepines, and it is possible formed in Italy reported that SUD patients are more fre- that the COVID-19 measures and the worrying news quently tested for COVID-19 infection, but the probabil- about the consequences of the pandemic have increased ity of a positive result is lower than in the general the risk of using anxiolytics to face emotional distress. population [28]. It should be mentioned that during the Therefore, it is still of interest to work on coping strategies study period COVID-19, tests were still widely unavail- in patients with BUD; and it should be noted that in Spain, able; therefore, it is difficult to make comparisons. benzodiazepines require a prescription, so most of these This study found that about 40% of the patients wors- patients were probably seen by a specialist. In any case, ened their consumption pattern, increasing quantity and benzodiazepine use should be constantly supervised in frequency of use during the lockdown established during facilities where SUD patients are treated [38]. the first months of the COVID-19 pandemic. Due to the Alcohol was expected to be the most consumed sub- design of the study, it is not possible to establish causal stance because alcohol use disorders are the most com- relationships regarding the impact of confinement; how- mon SUDs addressed in consultations and as its accessi- ever, it is of interest to compare the results with pre-pan- bility was not altered in the supermarkets during confine- demic studies focused on the evolution of SUD patients. ment. It is likely that some patients transferred their Studies carried out with outpatients, by participating cen- drinking in bars and restaurants to their homes, including tres in this study, showed that after 3 months of treat- drinking socially through digital channels; however, pa- ment, 44.2% of the patients had relapsed [32], and 80.5% tients who reduced their consumption could have taken had relapsed after one year [33]. As 60% of the patients in advantage of the confinement to avoid consumption. this study consumed some legal and/or illegal substances These data coincide with other Spanish studies carried during lockdown, it is possible to hypothesize that the out in SUD patients during this time [5, 20]. It also aligns confinement had a negative impact on the results of treat- with some studies carried out in the general population ment. The stress produced by the health crisis, the diffi- that found increased benzodiazepine and alcohol con- culties of continuing with face-to-face treatment, and sumption [9, 39]. However, other studies conducted in psychological distress may have affected treatment out- general population suggest alcohol consumption de- comes. Note that there are studies that question the effec- clined during the first months of the pandemic [40]. tiveness of psychosocial treatments for alcoholism and Regarding demographic characteristics, there were no that describe high relapse rates, with treatment outcomes differences in gender according to the consumption pat- affected by multiple factors [34, 35]. In addition, it is pos- tern during the period studied. However, differences sible that the patients who increased consumption had were observed when comparing based on age. It was ob- characteristics typical of addicted patients, such as diffi- served that younger patients more frequently reported culties in managing craving and other factors similar to having worsened consumption. In the multivariate analy- the general population that contributed to increased con- ses, age was maintained as a variable independently as- sumption, especially of alcohol, such as using substances sociated with greater severity. This finding contrasts with as coping strategies or due to boredom [36, 37] or because a study conducted on patients with alcohol use disorders they perceived fewer consequences of drinking more who attend a urine drug screening programme in Catalo- [15]. These motivational aspects could be integrated as nia which identified that the likelihood of testing posi- part of the psychological treatment. tively increased with age [20]. However, younger age has The substances that were most consumed by patients been associated with more impulsivity, and some studies were alcohol, cannabis, and cocaine. However, benzodi- have described poorer adherence and worse treatment azepines, cannabis, and tobacco were the substances of outcomes [32, 40]. Additionally, during the pandemic, 8 Eur Addict Res Grau-López et al. DOI: 10.1159/000521425 older patients may have been more careful as they identi- considered in the treatment offered to each patient. These fied themselves as part of the population more vulnerable issues illustrate the complexity of implementing telemed- to COVID-19. icine with addicted patients and highlight the need for When consumption pattern during lockdown was clinical guidelines on this issue [46]. compared in terms of previous SUD and dual diagnosis, Positive evaluation of the treatment was more frequent it was observed in the multivariate analysis that polysub- among patients who improved consumption. A favour- stance users and cocaine-dependent patients had wors- able evaluation of treatment had a significant positive re- ened consumption during lockdown. It is hypothesized lationship with drug use improvement outcomes [47]. that the increase in consumption in patients with these However, no study to date has examined changes in treat- SUDs may be associated with their tendency to have a ment patterns of individuals with SUD [27, 48], and it is high severity of addiction and the fact that they frequent- expected that further research will provide information ly have other psychiatric disorders [41]. Furthermore, in about the pros and cons of telemedicine in SUD patients. patients with SUD, it has been described at a neurobio- Patients in the worsened-pattern group more fre- logical level that stress is a critical determinant of relapse quently reported breaking the lockdown restrictions. susceptibility, especially in subjects with a history of high- This may be related with a greater severity of addiction, er frequency cocaine use [42, 43]. Therefore, stress coping antisocial behaviours, impulsivity, less self-control, and strategies should be included, especially in the treatment more emotional stress [49]. On the other hand, family of cocaine SUD patients. conflicts and unresolved social matters were indepen- Polysubstance use was more frequent in the worsened- dently associated with a poorer evolution of substance pattern group. Patients with 3 or more SUDs have a worse consumption. This relationship has been previously de- prognosis for the course of addiction [14], and in stressful scribed [50], and during the lockdown, as previous stud- situations, these people are likely to use substances to al- ies reported, social and familial problems have arisen [28, leviate emotional distress caused by home confinement. 51]. Regarding consumption patterns compared in terms The early phase of the COVID-19 pandemic has been of COVID-19-related variables during lockdown, it was related to an increase in depressive and anxious symp- observed at a multivariate level that telematic treatment toms [7, 10]. However, no all populations have shown this was more frequent in the maintained-pattern group in impact, as an Irish study has shown [52]. Our results have comparison to the other groups. Further research is re- shown relevant psychiatric distress in patients during quired to interpret these results and the role of telemedi- confinement, with frequent symptoms of anxiety, observ- cine in addicted patients, especially in times of social and ing that 40% of their general psychiatric status worsened. health crises. However, in this case, there was an associa- These data are consistent with other studies that have de- tion between face-to-face treatment and both improve- scribed mental health problems during the pandemic ment and worsening groups. It is possible that some pa- [53]. In addition, a relationship was observed between tients had achieved improvements when undergoing in- psychiatric symptoms and increased consumption dur- person treatment, and on the other hand, it is probable ing confinement. In the multivariate analysis, a higher that patients with a worse evolution had been visited in frequency of insomnia, more anxiety symptoms, and person, the telematic treatment being insufficient. Both greater symptoms measured by the general psychiatric mental health services and addiction treatment centres scale were described in the worsened-pattern group. This had to adapt to telemedicine or teletherapy during the result coincides with a study carried out in the general pandemic, e.g., virtual contingency management and also population that points out an increase in alcohol con- online group interventions [44]. Furthermore, it is pos- sumption in patients with more psychiatric difficulties sible that individual differences and the severity of the [39]. Moreover, this result reinforces the association be- addiction determine a patient’s capacity to adapt to tween mental distress and substance use. It is possible that telematic visits as these technologies may make it more substance consumption may have an anxiolytic effect and difficult to approach emotional aspects. Recently, the im- can in turn be a form of self-medication and of managing portance of digital inequalities has been raised, such as negative emotions, while conversely that substance use limited access to technology and lack of skills to leverage could also induce increased psychiatric symptoms. This the technology for desirable outcomes; these factors may highlights the need for comprehensive treatment that ad- contribute to negative health outcomes [45]. Digital in- dresses both psychiatric symptoms and substance use, equalities and individual features should be evaluated and even during a health crisis. COVID-19 Lockdown and Consumption Eur Addict Res 9 Patterns DOI: 10.1159/000521425 These results should be interpreted in light of their need to plan appropriate interventions in cases of similar limitations. Regrettably, we have not collected data from health crises is highlighted. patients who dropped out of treatment during the study, and it was not possible to perform any analysis on these patients. Our electronic medical record system dis- Acknowledgments charged 38.4% of the patients during the study period This study was coordinated by the Sub-Directorate General of because they did not request any consultation for more Drug Dependence of Health/Public Health Agency of Catalonia. than 6 months. Hence, the generalization of the results The study was also carried out thanks to involvement of the mem- may be restricted due to the limitations of the sample. A bers of the Xarxa d'Atenció a les Drogodependències – Grup de selection effect may be present as other studies declare Recerca sobre el COVID-19 (XAD-COVID-19 Group): Marta Perea Ortueta (Hospital Universitari Vall d’Hebron, Barcelona); that patients who reduce their substance consumption Nieves Martínez-Luna (Hospital Universitari Vall d’Hebron, Bar- are more likely to remain in treatment. The self-reported celona); Francina Fonseca (Hospital del Mar, Barcelona); Susana and retrospective nature of the data is another limitation Jornalé (CAS de la Mina, Barcelona); Xavier Samper (Hospital de because patients are less likely to admit substance con- Santa Maria, Lleida); Alicia Clariso (Hospital de Santa Maria, Llei- sumption to their therapists, or a memory bias could ex- da); Miriam Dueñas (CAS d’Horta-Guinardó, Barcelona); Ar- mando Tirado (CAS d’Horta-Guinardó, Barcelona); Begoña Irala ist [54]. Because of the COVID-19 measures, centres Indart (CAS Fontsanta, Cornellà de Llobregat); Rosa Pi i Marí conducted urine controls tests less frequently during (CAS Fontsanta, Cornellà de Llobregat); Víctor Martí Carrasco lockdown. Telematic treatment was not protocolized in (CAS de Rubí, Rubí); Amalia Gual de Torrella (CAS de Rubí, all centres, so future studies should assess the efficacy of Rubí); Francesco Panicali (CAS de Granollers, Granollers); Eu- these interventions. Moreover, unfortunately, it was not làlia Sabater Puig (Servei d’Addiccions i Salut Mental HUSJR, Reus); Tre Borràs i Cabacés (Servei d’Addiccions i Salut Mental possible to provide data from a control group; therefore, HUSJR, Reus); Silvia Martínez (CAS del Prat, El Prat del Llobre- it is not possible to make causal or explanatory interpre- gat); Beth Pallejà i Martos (CAS del Prat, El Prat del Llobregat); tations from the results. It has not been possible to isolate Raúl Yuste (CAS del Baix Llobregat, Sant Feliu de LLobregat); the effect of confinement from other variables, such as Carolina Franco (CAS de Les Corts, Barcelona); and Cinta Soler the evolution of patients treated for SUD. It is only pos- (CAS de Tortosa, Tortosa). sible to analyse the data at a descriptive level and to com- pare patient features according to the changes in their Statement of Ethics consumption pattern. On the other hand, the strengths of this study include This study was approved by the Ethics Committee of Vall that it is a multicentre study where thirteen centres from d’Hebron Hospital (PR-(AG)386–2020). Additionally, the legal Catalonia collaborated; it is important to highlight that department of the Sub-Directorate General of Drug Dependence there were no differences regarding consumption pat- of Health/Public Health Agency of Catalonia approved the data- base. Written informed consent was obtained from all partici- terns among urban versus non-urban centres. The study pants. included a large naturalistic sample, and the assessments were conducted by trained psychologists and psychia- trists specialized in addictions. Another strength of the Conflict of Interest Statement study is that it was carried out during a critical period at a social and health level. The information provided by Dr. Palma-Alvarez has received speaker honorariums from this research will allow the development of follow-up Angelini, Exeltis, Lundbeck, MSD, Mundipharma, and Takeda. Dr. Ramos-Quiroga has received speaker honorariums from Jans- studies during the evolution of the pandemic. sen-Cilag, Shire, Lilly, Ferrer, Medice, and Rubió. He has also re- We conclude that the home lockdown during the CO- ceived research funding from Janssen-Cilag, Lilly, Ferrer, Lund- VID-19 pandemic had major implications for substance beck, and Rubió. Dr. Grau-López has received speaker honorari- use and psychiatric distress among SUD outpatients. ums from Janssen-Cilag, Lundbeck, Servier, Otsuka, and Pfizer. More than 60% of the patients included in this study All other authors report no conflicts of interest and no financial relationships with commercial interests. maintained or worsened their consumption patterns during confinement, and only 38% improved through- out this time. Diverse factors were associated with the Funding Sources changes in the pattern like age, addiction severity, psy- chological distress during lockdown, social and familial The authors did not receive any funding in relation to this issues, and therapeutic variables. Considering this, the study. 10 Eur Addict Res Grau-López et al. DOI: 10.1159/000521425 Author Contributions Appendix Lara Grau-López contributed to conceptualization, methodol- Xarxa d’Atenció a les Drogodependències – Grup de Recerca ogy, validation, investigation, resources, data curation, writing sobre el COVID-19 (XAD-COVID-19-Group): Marta Perea Ortu- original draft, and visualization. Constanza Daigre contributed to eta (Hospital Universitari Vall d’Hebron, Barcelona); Nieves Mar- conceptualization, methodology, validation, investigation, formal tínez-Luna (Hospital Universitari Vall d’Hebron, Barcelona); analysis, resources, data curation, writing original draft, and visu- Francina Fonseca (Hospital del Mar, Barcelona); Susana Jornalé alization. Raúl Felipe Palma-Alvarez contributed to investigation, (CAS de la Mina, Barcelona); Xavier Samper (Hospital de Santa writing original draft, and data curation. Lidia Segura contributed Maria, Lleida); Alicia Clariso (Hospital de Santa Maria, Lleida); to conceptualization, methodology, resources, validation, writing Miriam Dueñas (CAS d’Horta-Guinardó, Barcelona); Armando original draft, and visualization. Marta Coronado contributed to Tirado (CAS d’Horta-Guinardó, Barcelona); Begoña Irala Indart conceptualization, methodology, resources, validation, writing (CAS Fontsanta, Cornellà de Llobregat); Rosa Pi i Marí (CAS Font- original draft, and visualization. Josep Antoni Ramos-Quiroga santa, Cornellà de Llobregat); Víctor Martí Carrasco (CAS de contributed to writing review and editing. Joan Colom contributed Rubí, Rubí); Amalia Gual de Torrella (CAS de Rubí, Rubí); Fran- to conceptualization, writing review, and editing. cesco Panicali (CAS de Granollers, Granollers); Eulàlia Sabater Puig (Servei d’Addiccions i Salut Mental HUSJR, Reus); Tre Borràs i Cabacés (Servei d’Addiccions i Salut Mental HUSJR, Reus); Silvia Martínez (CAS del Prat, El Prat del Llobregat); Beth Pallejà i Mar- Data Availability Statement tos (CAS del Prat, El Prat del Llobregat); Raúl Yuste (CAS del Baix Llobregat, Sant Feliu de LLobregat); Carolina Franco (CAS de Les The datasets made for this study can be available on demand to Corts, Barcelona); Cinta Soler (CAS de Tortosa, Tortosa). the corresponding author. Part of the data could not be allowed for distribution to others than the investigation group as it may violate ethical and/or legal regulations of written consent. References 1 World Drug Report 2020 [Internet]. United 8 Garre-Olmo J, Turró-Garriga O, Martí-Lluch 15 Bramness JG, Bye EK, Moan IS, Rossow I. Al- Nations publication [cited 2021 may 31]. R, Zacarías-Pons L, Alves-Cabratosa L, Serra- cohol use during the COVID-19 pandemic: Available from: https: //wdr.unodc.org/ no-Sarbosa D, et al. Changes in lifestyle re- self-reported changes and motives for change. wdr2020/field/WDR20_BOOKLET_1.pdf. sulting from confinement due to COVID-19 Eur Addict Res. 2021;27(4):257–62. 2 Legido-Quigley H, Mateos-García JT, Cam- and depressive symptomatology: a cross-sec- 16 Vanderbruggen N, Matthys F, Van Laere S, pos VR, Gea-Sánchez M, Muntaner C, McKee tional a population-based study. Compr Psy- Zeeuws D, Santermans L, Van den Ameele S, M. The resilience of the Spanish health system chiatry. 2021;104:152214. et al. Self-reported alcohol, Tobacco, and can- against the COVID-19 pandemic. Lancet 9 Jané-Llopis E, Anderson P, Segura L, Zabaleta nabis use during COVID-19 lockdown mea- Public Health. 2020;5(5):e251–e2. E, Muñoz R, Ruiz G, et al. Mental ill-health sures: results from a web-based survey. Eur 3 Deacon RM, Mammen K, Holmes J, Dunlop during COVID-19 confinement. BMC Psy- Addict Res. 2020;26(6):309–15. A, Bruno R, Mills L, et al. Assessing the valid- chiatry. 2021;21(1):194. 17 Koopmann A, Georgiadou E, Reinhard I, ity of the Australian treatment outcomes pro- 10 Vieta E, Pérez V, Arango C. Psychiatry in the Müller A, Lemenager T, Kiefer F, et al. The file for telephone administration in drug aftermath of COVID-19. Rev Psiquiatr Salud effects of the lockdown during the COVID-19 health treatment populations. Drug Alcohol Ment. 2020;13(2):105–10. pandemic on alcohol and tobacco consump- Rev. 2020;39(5):441–6. 11 Yao H, Chen JH, Xu YF. Patients with mental tion behavior in Germany. Eur Addict Res. 4 Green TC, Bratberg J, Finnell DS. Opioid use health disorders in the COVID-19 epidemic. 2021;27(4):242–56. disorder and the COVID 19 pandemic: a call Lancet Psychiatry. 2020;7(4):e21. 18 Lau JT, Yang X, Pang E, Tsui HY, Wong E, to sustain regulatory easements and further 12 Morisano D, Babor TF, Robaina KA. Co-oc- Wing YK. SARS-related perceptions in Hong expand access to treatment. Subst Abus. 2020; currence of substance use disorders with oth- Kong. Emerg Infect Dis. 2005;11(3):417–24. 41(2):147–9. er psychiatric disorders: Implications for 19 Kar SK, Arafat SMY, Sharma P, Dixit A, Mar- 5 Roncero C, Vicente-Hernández B, Casado- treatment services. Nordic Stud Alcohol thoenis M, Kabir R. COVID-19 pandemic and Espada NM, Aguilar L, Gamonal-Limcaoco S, Drugs. 2014;31(1):5–25. addiction: current problems and future con- Garzón MA, et al. The impact of COVID-19 13 Daigre C, Perea-Ortueta M, Berenguer M, cerns. Asian J Psychiatr. 2020;51:102064. pandemic on the castile and Leon addiction Esculies O, Sorribes-Puertas M, Palma-Alva- 20 Barrio P, Baldaquí N, Andreu M, Kilian C, treatment network: a real-word experience. rez R, et al. Psychiatric factors affecting recov- Rehm J, Gual A, et al. Abstinence among al- Front Psychiatry. 2020;11:575755. ery after a long term treatment program for cohol use disorder patients during the CO- 6 Samuels EA, Clark SA, Wunsch C, Jordison substance use disorder. Psychiatry Res. 2019; VID-19 pandemic: insights from Spain. Alco- Keeler LA, Reddy N, et al. Innovation during 276:283–9. hol Clin Exp Res. 2021;45(4):802–7. COVID-19: improving addiction treatment 14 Grau-López L, Roncero C, Daigre C, Gonzal- 21 Gili A, Bacci M, Aroni K, Nicoletti A, Gambe- access. J Addict Med. 2020;14(4):e8–e9. vo B, Bachiller D, Rodriguez-Cintas L, et al. lunghe A, Mercurio I, et al. Changes in drug 7 Brooks SK, Webster RK, Smith LE, Woodland [Risk factors for relapse in drug-dependent use patterns during the COVID-19 pandemic L, Wessely S, Greenberg N, et al. The psycho- patients after hospital detoxification]. Adic- in Italy: monitoring a vulnerable group by logical impact of quarantine and how to re- ciones. 2012;24(2):115–22. hair analysis. Int J Environ Res Public Health. duce it: rapid review of the evidence. Lancet. 2021;18(4):1967. 2020;395(10227):912–20. COVID-19 Lockdown and Consumption Eur Addict Res 11 Patterns DOI: 10.1159/000521425 22 Taylor S, Paluszek MM, Rachor GS, McKay to addiction severity and psychiatry comorbid- review. Brain Res Brain Res Rev. 2000; 33(1): D, Asmundson GJG. Substance use and ity: a six-month follow-up study in an outpa- 13–33. abuse, COVID-19-related distress, and disre- tient unit. Addict Behav. 2021;117:106832. 44 Zastepa E, Sun JC, Clune J, Mathew N. Adapta- gard for social distancing: a network analysis. 33 Palma-Álvarez RF, Ros-Cucurull E, Daigre C, tion of contingency management for stimulant Addict Behav. 2021;114:106754. Perea-Ortueta M, Martínez-Luna N, Serrano- use disorder during the COVID-19 pandemic. 23 Shreffler J, Petrey J, Huecker M. The impact Pérez P, et al. Is alexithymia related to reten- J Subst Abuse Treat. 2020;118:108102. of COVID-19 on healthcare worker wellness: tion and relapses in patients with substance 45 Garett R, Young SD. Potential effects of digi- a scoping review. West J Emerg Med. 2020; use disorders?: A one year follow-up study. tal inequality on treatment seeking for opioid 21(5):1059–66. Addict Behav. 2021;113:106681. use disorder. Int J Ment Health Addict. 2021: 24 Dunlop A, Lokuge B, Masters D, Sequeira M, 34 Cutler RB, Fishbain DA. Are alcoholism 27:1–6. Saul P, Dunlop G, et al. Challenges in main- treatments effective? The project MATCH 46 Barney A, Buckelew S, Mesheriakova V, Ray- taining treatment services for people who use data. BMC Public Health. 2005;5:75. mond-Flesch M. The COVID-19 pandemic drugs during the COVID-19 pandemic. 35 UKATT Research Team. Effectiveness of and rapid implementation of adolescent and Harm Reduct J. 2020;17(1):26. treatment for alcohol problems: findings of young adult telemedicine: challenges and op- 25 Picchio CA, Valencia J, Doran J, Swan T, Pas- the randomised UK alcohol treatment trial portunities for innovation. J Adolesc Health. tor M, Martró E, et al. The impact of the CO- (UKATT). BMJ. 2005;331(7516):541. 2020;67(2):164–71. VID-19 pandemic on harm reduction servic- 36 Graupensperger S, Fleming CB, Jaffe AE, 47 Zhang Z, Gerstein DR, Friedmann PD. Pa- es in Spain. Harm Reduct J. 2020;17(1):87. Rhew IC, Patrick ME, Lee CM. Changes in tient satisfaction and sustained outcomes of 26 Vallecillo G, Anguera M, Martin N, Robles young adults’ alcohol and marijuana use, drug abuse treatment. J Health Psychol. 2008; MJ. Effectiveness of an acute care for elders norms, and motives from before to during the 13(3):388–400. unit at a long-term care facility for frail older COVID-19 pPandemic. J Adolesc Health. 48 Blanco C, Compton WM, Volkow ND. Op- patients with COVID-19. Geriatr Nurs. 2021 2021;68(4):658–65. portunities for research on the treatment of Mar-Apr;42(2):544–7. 37 Vedelago L, Wardell JD, Kempe T, Patel H, substance use disorders in the context of CO- 27 Volkow ND. Collision of the COVID-19 and Amlung M, MacKillop J, et al. Getting high to VID-19. JAMA Psychiatry. 2021;78(4):357–8. addiction epidemics. Ann Intern Med. 2020; cope with COVID-19: Modelling the associa- 49 Nivette A, Ribeaud D, Murray A, Steinhoff A, 173(1):61–2. tions between cannabis demand, coping mo- Bechtiger L, Hepp U, et al. Non-compliance 28 Djuric O, Mancuso P, Zannini A, Nicolaci A, tives, and cannabis use and problems. Addict with COVID-19-related public health mea- Massari M, Zerbini A, et al. Are individuals Behav. 2021;124:107092. sures among young adults in Switzerland: in- with substance use disorders at higher risk of 38 Soyka M. Treatment of Benzodiazepine De- sights from a longitudinal cohort study. Soc SARS-CoV-2 infection? population-based pendence. N Engl J Med. 2017;376(12):1147– Sci Med. 2021;268:113370. registry study in Northern Italy. Eur Addict 57. 50 Daley DC. Family and social aspects of sub- Res. 2021;27(4):263–7. 39 Smith BJ, Lim MH. How the COVID-19 pan- stance use disorders and treatment. J Food 29 Guy W. ECDEU assessment manual for psy- demic is focusing attention on loneliness and Drug Anal. 2013;21(4):S73–6. chopharmacology. Rockville: US Department social isolation. Public Health Res Pract. 2020; 51 Rodriguez-Jimenez R, Fares-Otero NE, Gar- of Health, Education, and Welfare, Public 30(2):3022008. cía-Fernández L. Gender-based violence dur- Health ServiceService Alcohol, Drug Abuse, 40 Brorson HH, Ajo Arnevik E, Rand-Hendrik- ing COVID-19 outbreak in Spain. Psychol and Mental Health Administration; 1976. sen K, Duckert F. Drop-out from addiction Med. 2020;7:1–5. 30 Bulbena A, Bobes J, Luque A, Dal-Ré R, Ball- treatment: a systematic review of risk factors. 52 Hyland P, Shevlin M, McBride O, Murphy J, esteros J, Ibarra N, et al. [A validation study of Clin Psychol Rev. 2013;33(8):1010–24. Karatzias T, Bentall RP, et al. Anxiety and de- the Spanish versions of the Clinical Anxiety 41 Breese GR, Sinha R, Heilig M. Chronic alco- pression in the republic of Ireland during the Scale and the Physician Questionnaire for the hol neuroadaptation and stress contribute to COVID-19 pandemic. Acta Psychiatr Scand. assessment of anxiety disorders]. Med Clin. susceptibility for alcohol craving and relapse. 2020;142(3):249–56. 2003;121(10):367–74. Spanish. Pharmacol Ther. 2011;129(2):149–71. 53 Vindegaard N, Benros ME. COVID-19 pan- 31 Overall JE, Hollister LE, Pichot P. Major psy- 42 Daigre C, Terán A, García-Vicent V, Roncero demic and mental health consequences: sys- chiatric disorders. a four-dimensional model. C. Attention deficit hyperactivity disorder tematic review of the current evidence. Brain Arch Gen Psychiatry. 1967;16(2):146–51. and central nervous system depressants de- Behav Immun. 2020;89:531–42. 32 Daigre C, Rodríguez L, Roncero C, Palma-Ál- pendence. A review. Adicciones. 2013; 25(2): 54 Del Boca FK, Darkes J. The validity of self- varez RF, Perea-Ortueta M, Sorribes-Puertas 171–86. reports of alcohol consumption: state of the M, et al. Treatment retention and abstinence of 43 Shaham Y, Erb S, Stewart J. Stress-induced re- science and challenges for research. Addic- patients with substance use disorders according lapse to heroin and cocaine seeking in rats: a tion. 2003;98(Suppl 2):1–12. 12 Eur Addict Res Grau-López et al. DOI: 10.1159/000521425