Search

Anxiety, depression, and co-morbid alcohol use disorder (AUD) are frequent and often difficult to diagnose and treat accurately. Diagnosis of anxiety or depression should only be made when the patients have been abstinent for four to six weeks. Patients are treated in general practice, municipal alcohol outpatient clinics, or in psychiatric settings. There are no national guidelines for this organisation in Denmark. Integrated treatment (psychotherapy and/or medical), at which both anxiety or depression and AUD are treated simultaneously by the same therapist (team), seems to be best.

The least trimmed squares (LTS) estimator is a popular robust regression estimator. It finds a subsample of h ‘good’ observations among n observations and applies least squares on that subsample. We formulate a model in which this estimator is maximum likelihood. The model has ‘outliers’ of a new type, where the outlying observations are drawn from a distribution with values outside the realized range of h ‘good’, normal observations. The LTS estimator is found to be h1/2 consistent and asymptotically standard normal in the location-scale case. Consistent estimation of h is discussed. The model differs from the commonly used ϵ-contamination models and opens the door for statistical discussion on contamination schemes, new methodological developments on tests for contamination as well as inferences based on the estimated good data.

Aims Many patients with alcohol use disorders are challenged by cravings leading to repeated relapses. Both cue exposure therapy (CET) and acamprosate target alcohol cravings and are often combined (CET + acamprosate). The main aim of this study was to investigate whether aftercare treatment consisting of CET combined with acamprosate is equivalent to (A) CET as monotherapy, (B) aftercare as usual (AAU) as monotherapy or (C) AAU combined with acamprosate. Methods Patients were randomized to receive either CET with urge-specific coping skills (USCS) as aftercare or AAU. Acamprosate prescription data were extracted from patient case records. Alcohol consumption, cravings, and USCS were assessed at pre-aftercare, post-aftercare, and 6-month follow-up. Results Overall, patients increased their alcohol consumption during and following aftercare treatment, thereby relapsing despite any treatment. However, CET + acamprosate achieved greater abstinence compared to AAU + acamprosate at follow-up (p=.047). CET + acamprosate also reduced number of drinking days (p=.020) and number of days with excessive drinking (p=.020) at post-aftercare, when compared to AAU monotherapy. CET monotherapy increased sensible drinking at post-aftercare compared to AAU monotherapy (p=.045) and AAU + acamprosate (p=.047). Only CET monotherapy showed improvement in cravings, when compared to AAU at follow-up (mean urge level: p=.032; peak urge level: p=.014). Conclusion The study showed that CET both as monotherapy and combined with acamprosate was superior to AAU monotherapy and AAU + acamprosate in reducing alcohol consumption. Only CET + acamprosate was capable of reducing alcohol consumption in the longer term, indicating that anti-craving medication may not impede CET from exerting an effect on alcohol consumption. Trial registration: ClinicalTrials.gov ID: NCT02298751 (24/11-2014).

Aim To investigate if more patients in the intervention group attended specialized alcohol treatment compared with a treatment-as-usual group (TAU). Methods Pragmatic randomized controlled trial where consecutive patients, admitted to somatic hospitals, filled out a lifestyle questionnaire with the Alcohol Use Disorder Identification Test (AUDIT) embedded. Patients scoring 8+ on AUDIT were included in the study. Included patients were randomized to either a Danish screening brief intervention and referral to treatment (SBIRT) called the Relay model or TAU depending on date of admission. The Relay group was offered a brief alcohol intervention by an outreach alcohol therapist. Patients scoring 16 points and above on the AUDIT test also received referral to alcohol treatment. Outcome was attendance at specialized outpatient alcohol treatment centres after discharge from hospital. Information on patients was gathered from municipal databases at 18 months follow-up. Results A total of 3534 patients completed the questionnaire, and 609 patients (17%) scored AUDIT 8+. 48 patients were lost to follow-up, and the final sample had 561 patients. Only 33 patients (6%) attended outpatient treatment at 18-months follow-up, but significantly more patients in the Relay group sought alcohol treatment than in the TAU group (OR = 2.5 [1.2;5.2] (p = 0.017)). Number needed to treat (NNT) was 20 [95% CI 11.2;112.3]. Conclusion The Relay intervention was associated with more patients attending specialized treatment, but further research is needed to establish if general hospitals are an excellent platform for performing SBIRT.

Faculty and graduates of the University of Oxford have played a significant role in the history of econometrics from an early date. The term econometrics was only formulated by Ragnar Frisch in the 1930s, but in the seventeenth century, William Petty created a discipline that he called Political Arithmetick, a forerunner of quantitative economics that led to the more specialised statistical approach of econometrics. During the first half of the twentieth century, Oxford scholars like Colin Clark made major advances in creating aggregate economic measurements. From the late 1970s, the focus was primarily on macroeconometrics for the remainder of that century, buttressed by research on methods for analysing dynamic panels. In the twenty-first century, micro-econometrics was added to the portfolio. The most recent addition is climate econometrics, developing and applying econometric tools for analysing climate data, which is driven by human economic behaviour and so faces much the same slew of problems as macroeconomic time series.

Introduction: The aim was to describe the treatment-seeking population with alcohol use disorder by means of data from the AUDIT questionnaire at referral and to evaluate and discuss if this information can be of use in treatment and service planning. Methods: Data from 2016 to 2018 were extracted from the National Database on Substance Abuse Treatment. The sample included 1281 individual treatment seekers from all over Greenland. Results: Mean age was 38.2 years (SD 12.1 years). Only 60.1% had a total AUDIT score suggesting dependency, and 15.5% had a harmful use. While most only drank 2–4 times a month, about 95% binge drank. Half reported loss of control at least weekly, and one of three had been unable to do what was expected of them, or needed a drink first thing in the morning weekly or more often. In two-third others had been concerned about the drinking. Users of cannabis had a higher AUDIT score, while gambling was unassociated to alcohol use. Substantial gender and regional differences were seen. Discussion: The AUDIT screening was found useful in individual and national treatment planning. Data suggested that particular focus should be given to women in treatment, and the service offered to the East Greenlandic population.

We develop an age-period-cohort model for repeated cross-section data with individual covariates, which identifies the non-linear effects of age, period and cohort. This is done for both continuous and binary dependent variables. The age, period and cohort effects in the model are represented by a parametrization with freely varying parameters that separates the identified non-linear effects and the unidentifiable linear effects. We develop a test of the parametrization against a more general ‘time-saturated’ model. The method is applied to analyse the obesity epidemic in England using survey data. The main non-linear effects we find in English obesity data are age-related among women and cohort-related among men.

Introduction: In 2016, a new addiction treatment service was established in Greenland to tackle the addiction problems with alcohol, cannabis and gambling among the population. The new service has established a treatment center in each of the five municipalities and works in partnership with a central private provider of treatment for those who reside in areas without a local treatment center.Methods: The national addiction database provided us with data from the Alcohol Use Disorder Identification Test, Alcohol Severity Index and questions on cannabis use and gambling behavior received at referral to, and at initiation of treatment. The data were analyzed for differences between the population in local or central treatment using SPSS version 25 (SPSS Inc., Chicago, IL).Results: Significant differences between the individuals in local and central treatment were revealed. Individuals in local treatment are more often women with minor children and a job, and their alcohol use is concentrated on weekends/holidays. Individuals in central treatment are more equal in both genders, few have minor children living at home, heavy drinking is more pronounced, and cannabis is used more frequently as well.Discussion: The findings support our expectations of local treatment being more attractive to individuals with obligations at home. The differences in the populations are worth considering when planning the treatment service, as the needs of the populations might differ. The findings are limited by many missing in the analyses, which we believe is caused by the establishing process of the new service

This study explores health and social care consumption in two groups of patients with risk of alcohol use disorder (AUD), following a brief outreach alcohol intervention in a general hospital setting in Denmark. The Relay intervention aims to decrease health care contacts and thus primarily, in the long run, to reduce health care costs and secondarily to reduce labour market consequences and social costs for patients with alcohol problems. The study took place in somatic hospital departments with high prevalence of alcohol related injuries and illnesses. Patients admitted to the hospital between October 2013 and June 2016 were screened using the Alcohol Use Identification Test (AUDIT) and everyone scoring 8 points and above were randomised to either intervention (Relay group) or control group (TAU group). The patients (n=561) were followed for 12 months after discharge from the hospital. Data was gathered on somatic and psychiatric hospital admissions, GP visits and other primary health care visits as well as the costs associated with the health care contacts. In addition, data on social costs and productivity was gathered. All data was gathered from the Danish registers using personal identification numbers. We modelled the association using generalised linear modelling and investigated the costs further by performing a quantile analysis. We found no statistically significant difference in health care costs, social costs or productivity between the two groups. A longer follow-up is needed to fully investigate effects of the Relay intervention on changes in patients’ health behaviour and subsequently on health care costs.

BackgroundSocial phobia (SP) is one of the most prevalent anxiety disorders among patients entering treatment for alcohol use disorders (AUD). However, few studies have examined the association between SP and alcohol-related treatment outcomes in naturalistic settings.AimsThe aims of this study were to estimate the prevalence of co-morbid SP and to investigate the impact of the co-morbidity on the treatment course, dropout rates and treatment outcomes in a large sample of AUD patients treated in an outpatient alcohol treatment clinic.MethodsThe study was conducted as an observational cohort study. A consecutive sample of 3.197 treatment-seeking outpatients, with an AUD diagnosis according to the ICD-10 Diagnostic Criteria for Research, was assessed by means of the Addiction Severity Index at treatment start and at treatment conclusion.ResultsApproximately 15% of the patients suffered from SP when entering treatment and patients with and without SP did not differ on the treatment course, compliance or dropout rates. SP did not predict any alcohol-related treatment outcomes either, where no association was found on change scores for abstinence, drinking days and days with excessive drinking relative to AUD patients without co-morbidity.ConclusionAUD patients with and without co-morbid SP were equally likely to achieve benefits when treated similarly with evidence-based pharmacological and psychosocial approaches in a naturalistic setting

Purpose: This study examined 6-month drinking outcomes of elderly patients compared with middle-aged patients in a clinical sample after initiation of outpatient treatment for alcoholism. Materials and methods: In a clinical prospective cohort study, 1398 consecutive patients from a municipality outpatient alcohol clinic were included. A total of 208 elderly patients aged from 60 to 82 years and 1190 middle-aged patients from 40 to 59 years participated in the study. The following psychosocial treatment interventions were offered: cognitive behavioral therapy, family therapy and supportive consultations. Using an ‘intention-to-treat’ method, primary outcomes included drinking outcomes (self-reported abstinence rates, drinking 3 drinks or less per day, and change in Addiction Severity Index [ASI] composite scores) during the 30 days prior to 6-month follow-up; secondary outcome was compliance to the recommended treatment. Results: Compared to middle-aged, among elderly patients a higher proportion were females (33.5% vs. 42.8%) and had a lower family/social ASI-composite score (0.17 vs. 0.12) at baseline. Higher alcohol and family/social ASI-composite scores were inversely correlated with abstinence. Elderly patients had a higher chance for abstinence compared to middle-aged patients (Odds ratio 95% [confidence interval]) 1.40 (1.03–1.92). The proportion of elderly patients that drank 3 or less drinks per day was 17.8%, compared to 10.8% among middle-aged (p

This study investigates the benefit of adding a data-driven actuarial matching system and focus on motivation to routine outpatient alcoholism treatment. The study is a two-cohort quasi-experimental study. One cohort (188 clients) were assigned to treatment based on clinician judgment. Cohort 2 (175 clients) were assigned to treatment based on actuarial matching procedures. Both cohorts were followed for 12 months. A total of 37% of Cohort 2 dropped out of treatment, compared to 52% of Cohort 1. Cohort 2 also showed greater reduction in alcohol consumption and greater improvement. Actuarial matching procedures appear to have improved treatment retention and outcome.

Background: In the present study, we investigated the overall lifestyles of patients with hazardous alcohol use and alcohol dependence who were admitted to the hospital and investigated unhealthy lifestyle factors and their clustering in inpatients. Methods: Patients admitted to the gastrointestinal, neurologic or orthopedic departments at Odense University Hospital or to the emergency department at Aabenraa Hospital in the inclusion period, October 2013 to June 2016, completed a lifestyle questionnaire asking questions about their diet, alcohol consumption, exercise and smoking habits. Patients were divided into three groups depending on their score from the alcohol use disorder identification test, which was embedded in the lifestyle questionnaire, and odds ratios were calculated using logistic regression. Results: Patients with alcohol dependence had statistically significantly higher odds of being smokers, having unhealthy diets and being physically inactive compared with patients without alcohol problems. Among patients with hazardous alcohol drinking, we found an increased occurrence of smokers and an inverse association between hazardous alcohol drinking and being physically inactive. Many of the patients had attempted to change their unhealthy lifestyles. Conclusion: We found that alcohol problems are related to a clustering of other lifestyle factors and that many of the patients admitted to certain departments showed signs of various kinds of alcohol problems. Therefore, specific hospital departments could be opportune settings for preventive alcohol interventions.

Aim: In 2007, the legal obligation to deliver alcohol treatment to the public was transferred to the 98 municipalities of Denmark. This resulted in changes in how alcohol treatment centers in Denmark work. The aim of the present study was to describe the patient profiles and treatment outcomes in the alcohol treatment centers regarding regional variation and changes over time. Subjects and methods: This is a descriptive, register-based study of patients enrolled in alcohol treatment centers from 2006–2014 in Denmark. Only patients above the age of 15 years and with a valid postal code were included. The sample was restricted to the patients’ first contact with the alcohol treatment register (n = 44,516). Results: Patients who initiated treatment in the period 2006–2014 and were registered in the National Alcohol Treatment Register were primarily males (69–70%) with an average age between 46 and 49 years. Most had a vocational education (38–41%) on top of their primary education. The number of years with excessive alcohol use started out as being quite different in the five regions, but became more homogeneous over the study period. Treatment duration in the various regions followed a similar pattern, with all five of them having a similar treatment duration time of 160–230 days by 2014. Conclusion: We found that treatment for alcohol use disorder became more homogeneous across the regions in Denmark over time and that by 2014 it was difficult to identify any differences across the country.

The Least Trimmed Squares (LTS) and Least Median of Squares (LMS) estimators are popular robust regression estimators. The idea behind the estimators is to find, for a given h, a sub-sample of h `good' observations among n observations and estimate the regression on that sub-sample. We find models, based on the normal or the uniform distribution respectively, in which these estimators are maximum likelihood. We provide an asymptotic theory for the location-scale case in those models. The LTS estimator is found to be sqrt(h) consistent and asymptotically standard normal. The LMS estimator is found to be h consistent and asymptotically Laplace.

We show that the cumulated sum of squares statistic has a standard Brownian bridge–type asymptotic distribution in nonlinear regression models with (possibly) nonstationary regressors. This contrasts with cumulated sum statistics which have been previously studied and whose asymptotic distribution has been shown to depend on the functional form and the stochastic properties, such as persistence and stationarity, of the regressors. A recursive version of the test is also considered. A local power analysis is provided, and through simulations, we show that the test has good size and power properties across a variety of situations.

Introduction A paternalistic approach to treatment planning is common, but there is a growing interest in involving clients in the planning of their own treatment, both in medical care and psychotherapy. Several studies on matching patients to treatment have failed to improve outcome. Studies on Shared Decision-Making in mental health are encouraging but somewhat ambiguous in addiction treatment. Objective This study investigated whether self-matching (Informed Choice) to treatment improves alcohol consumption, retention, and quality of life. Method 402 consecutive clients (female 46.7 %, mean age 47.4) seeking treatment for alcohol use disorder at the outpatient clinic in Odense, Denmark were enrolled in this single-blinded randomized controlled trial. Clients randomized to the intervention group watched an 8-minute video presentation of the treatments and then chose the treatment they preferred. Clients in the treatment as usual (TAU) group were allocated to treatment by an evidence-based algorithm. Measurements on alcohol consumption, quality of life, and retention in care were completed at baseline and 6-month follow-up. Results There was no significant difference on primary and secondary outcome measures between the Informed Choice group and TAU group at 6-month follow-up. 80 % of clients in the Informed Choice group were satisfied with being randomized to self-matching whereas 24 % in the TAU group were satisfied with being assigned by expert matching. Conclusion Our hypothesis, that outcome is improved if clients choose their own treatment, was not supported. However, an important finding is that client self-matching is just as beneficial as expert matching.

Outlier detection algorithms are intimately connected with robust statistics that down-weight some observations to zero. We define a number of outlier detection algorithms related to the Huber-skip and least trimmed squares estimators, including the one-step Huber-skip estimator and the forward search. Next, we review a recently developed asymptotic theory of these. Finally, we analyse the gauge, the fraction of wrongly detected outliers, for a number of outlier detection algorithms and establish an asymptotic normal and a Poisson theory for the gauge.

BACKGROUND Cue exposure therapy (CET) is a psychological approach developed to prepare individuals with alcohol use disorder (AUD) for confronting alcohol and associated stimuli in real life. CET has shown promise when treating AUD in group sessions, but it is unknown whether progressing from group sessions to using a mobile phone app is an effective delivery pathway. OBJECTIVE The objectives of this study were to investigate (1) whether CET as aftercare would increase the effectiveness of primary treatment with cognitive behavior therapy, and (2) whether CET delivered through a mobile phone app would be similarly effective to CET via group sessions. METHODS A total of 164 individuals with AUD were randomized to one of three groups: CET as group aftercare (CET group), CET as fully automated mobile phone app aftercare (CET app), or aftercare as usual. Study outcomes were assessed face-to-face at preaftercare, postaftercare, and again at 6 months after aftercare treatment. Generalized mixed models were used to compare the trajectories of the groups over time on drinking, cravings, and use of urge-specific coping skills (USCS). RESULTS In all, 153 of 164 individuals (93%) completed assessments both at posttreatment and 6-month follow-up assessments. No differences in the trajectories of predicted means were found between the experimental groups (CET group and app) compared with aftercare as usual on drinking and craving outcomes over time. Both CET group (predicted mean difference 5.99, SE 2.59, z=2.31, P=.02) and the CET app (predicted mean difference 4.90, SE 2.26, z=2.31, P=.02) showed increased use of USCS compared to aftercare as usual at posttreatment, but this effect was reduced at the 6-month follow-up. No differences were detected between the two experimental CET groups on any outcomes. CONCLUSIONS CET with USCS delivered as aftercare either via group sessions or a mobile phone app did not increase the effectiveness of primary treatment. This suggests that CET with USCS may not be an effective psychological approach for the aftercare of individuals treated for AUD. CLINICALTRIAL ClinicalTrials.gov NCT02298751; https://clinicaltrials.gov/ct2/show/NCT02298751

An extended and improved theory is presented for marked and weighted empirical processes of residuals of time series regressions. The theory is motivated by 1-step Huber-skip estimators, where a set of good observations are selected using an initial estimator and an updated estimator is found by applying least squares to the selected observations. In this case, the weights and marks represent powers of the regressors and the regression errors, respectively. The inclusion of marks is a non-trivial extention to previous theory and requires refined martingale arguments.

A uniform weak consistency theory is presented for the marked and weighted empirical distribution function of residuals. New and weaker sufficient conditions for uniform consistency are derived. The theory allows for a wide variety of regressors and error distributions. We apply the theory to 1-step Huber-skip estimators. These estimators describe the widespread practice of removing outlying observations from an initial estimation of the model of interest and updating the estimation in a second step by applying least squares to the selected observations. Two results are presented. First, we give new and weaker conditions for consistency of the estimators. Second, we analyze the gauge, which is the rate of false detection of outliers, and which can be used to decide the cut-off in the rule for selecting outliers.

We propose an asymptotic theory for distribution forecasting from the log normal chain-ladder model. The theory overcomes the difficulty of convoluting log normal variables and takes estimation error into account. The results differ from that of the over-dispersed Poisson model and from the chain-ladder based bootstrap. We embed the log normal chain-ladder model in a class of infinitely divisible distributions called the generalized log normal chain-ladder model. The asymptotic theory uses small $\sigma$ asymptotics where the dimension of the reserving triangle is kept fixed while the standard deviation is assumed to decrease. The resulting asymptotic forecast distributions follow t distributions. The theory is supported by simulations and an empirical application.

This is an advance summary of a forthcoming article in the Oxford Research Encyclopedia of Economics and Finance. Please check back later for the full article. Detection of outliers is an important explorative step in empirical analysis. Once detected, the investigator will have to decide how to model the outliers depending on the context. Indeed, the outliers may represent noisy observations that are best left out of the analysis or they may be very informative observations that would have a particularly important role in the analysis. For regression analysis in time series a number of outlier algorithms are available, including impulse indicator saturation and methods from robust statistics. The algorithms are complex and their statistical properties are not fully understood. Extensive simulation studies have been made, but the formal theory is lacking. Some progress has been made toward an asymptotic theory of the algorithms. A number of asymptotic results are already available building on empirical process theory.

We show boundedness in probability uniformly in sample size of a general M-estimator for multiple linear regression in time series. The positive criterion function for the M-estimator is assumed lower semicontinuous and sufficiently large for large argument. Particular cases are the Huber-skip and quantile regression. Boundedness requires an assumption on the frequency of small regressors. We show that this is satisfied for a variety of deterministic and stochastic regressors, including stationary and random walks regressors. The results are obtained using a detailed analysis of the condition on the regressors combined with some recent martingale results.

Purpose: Many patients with alcohol use disorder (AUD) fail to initiate outpatient treatment after being discharged from inpatient detoxification. The aim of the present study was to investigate the effect of outreach visits (OV) on subsequent attendance in outpatient AUD treatment.Methods: Patients admitted to a detoxification unit at a mental health hospital were randomized to either referral as usual (TAU) or OV. OV involved a nurse from the outpatient clinic meeting the patients at the hospital and advising them about outpatient treatment, presenting an ‘attendance contract’ and offering a first appointment.Results: One hundred consecutive patients were randomized while hospitalized, 55 received TAU and 45 OV. About 84.4% OV patients presented for treatment at the outpatient clinic compared to 58.2% TAU-patients (p = .004). About 46.7% OV patients received continuing care for at least 3 months compared to 23.6% TAU-patients (p = .016).Conclusion: OV are an effective strategy for ensuring continued treatment.

The classical Chow test for structural instability requires strictly exogenous regressors and a break-point specified in advance. In this paper, we consider two generalisations, the one-step recursive Chow test (based on the sequence of studentised recursive residuals) and its supremum counterpart, which relaxes these requirements. We use results on the strong consistency of regression estimators to show that the one-step test is appropriate for stationary, unit root or explosive processes modelled in the autoregressive distributed lags (ADL) framework. We then use the results in extreme value theory to develop a new supremum version of the test, suitable for formal testing of structural instability with an unknown break-point. The test assumes the normality of errors and is intended to be used in situations where this can be either assumed nor established empirically. Simulations show that the supremum test has desirable power properties, in particular against level shifts late in the sample and against outliers. An application to U.K. GDP data is given.

Cue Exposure Therapy (CET) is a behavioristic psychological approach to treating substance use disorders (SUD). Prior systematic reviews have found CET to be ineffective when targeting SUDs. The effect of this approach on alcohol use disorders (AUD) seems more promising at trial level but has yet to be systematically reviewed and quantitatively analyzed. Therefore, we aimed to examine the effectiveness of CET targeting AUD compared to active control conditions in a meta-analytic review. Following a systematic search of the literature, a total of seven controlled trials were identified. CET showed no to small additional effects on drinking intensity and drinking frequency, a small additional effect on total drinking score and a moderate additional effect on latency to relapse. Stratification and analysis of a-priori defined trial covariates revealed that CET may have an increased effect in the longer term, and that CET combined with urge-specific coping skills may be the better option for treating AUD than conventional CET. Also, CET may prove less effective when comparing it to cognitive behaviour therapy as opposed to other active control conditions. The overall quality of evidence was graded low due to high risk of bias, inconsistency, imprecision and suspected publication bias. Sounder methodological trials are needed to derive a firm conclusion about the effectiveness of CET for treating AUD.

This article investigates the original Chinese sources for two circular trigram arrangements that have played a crucial role in numerology associated with the Book of Changes (the Yijing 《易 經》) since the Song dynasty (960–1279). While attempting to clarify the nature and the origins of these and related diagrams, recent secondary literature on topic is reviewed, especially the influential articles by Schuyler Cammann.

A new Bornhuetter–Ferguson method is suggested herein. This is a variant of the traditional chain ladder method. The actuary can adjust the relative ultimates using externally estimated relative ultimates. These correspond to linear constraints on the Poisson likelihood underpinning the chain ladder method. Adjusted cash flow estimates were obtained as constrained maximum likelihood estimates. The statistical derivation of the new method is provided in the generalised linear model framework. A related approach in the literature, combining unconstrained and constrained maximum likelihood estimates, is presented in the same framework and compared theoretically. A data illustration is described using a motor portfolio from a Greek insurer., This research was funded by the programme for Economic Modelling, Oxford and European Research Council, grant AdG 694262, and the Spanish Ministry of Economy and Competitiveness, grant MTM2016-76969P, which includes support from the European Regional Development Fund (ERDF).

Summary It is of considerable interest to forecast future mesothelioma mortality. No measures for exposure are available so it is not straightforward to apply a dose–response model. It is proposed to model the counts of deaths directly by using a Poisson regression with an age–period–cohort structure, but without offset. Traditionally the age–period–cohort is viewed as suffering from an identification problem. It is shown how to reparameterize the model in terms of freely varying parameters, to avoid this problem. It is shown how to conduct inference and how to construct distribution forecasts.