Unrecorded alcohol consumption may be a global problem, with about 25% of all alcohol consumption concerning this category. There are different styles of unrecorded alcohol, legally produced versus illegally produced, artisanal vs industrially produced, and so surrogate alcohol, which is officially not intended for human consumption. Monitoring and surveillance of unrecorded consumption aren’t well developed. the planet Health Organization has developed a monitoring system, using the Nominal Group Technique, a variant of the Delphi methodology. Experiences with this technique over the past two years are reported. Finally, conclusions for the monitoring and surveillance at the national level are given.
Unrecorded consumption: definition, categories, and estimated size
Unrecorded consumption, concerning alcohol that’s outside the standard system of governmental control because it’s produced, distributed, and sold outside of formal channels and, therefore, not registered by routine data collection , , represented about 25% of all the consumption globally within the year 2012 . The relative proportion is way higher in low-income than in high-income countries : of course, the relative proportion of unrecorded consumption increases almost proportionally with decreasing wealth (high-income countries: 9%; upper-middle-income countries: 24%; lower-middle-income countries: 42%; lower-income countries: 44%). the explanations for this relation are a minimum of twofold: first, there’s more informal traditional fermented beverages and spirits production in low- and middle-income countries , , irrespectively if such artisanal production is legal, tolerated, or illegal; and second, the enforcement of taxation rules is lower and corruption is higher in less wealthy countries . However, while this relationship between economic wealth and unrecorded consumption exists on a rustic level supported current data, potential measurement bias in volumes of unrecorded consumption should be acknowledged despite attempts of the planet Health Organization to cut back this bias , .
As a results of the high prevalence of unrecorded alcohol products, monitoring and surveillance efforts of alcohol as a significant risk factor for health , ,  have to consider consumption of unrecorded alcohol, to both estimate consumption and alcohol-attributable harms, and to judge policy measures. Many countries have an interest in knowing what proportion of overall alcohol consumption stems from unrecorded sources, and also the composition of unrecorded differs widely by culture. Unrecorded alcohol comprises the subsequent categories , , , :
•Alcohol that’s originally not produced for human consumption (such as medicinal products that contain alcohol, perfume, or industrial alcohol); in some countries (e.g., Russia), such surrogate alcohol may only officially be declared as not for human consumption to avoid taxes. this can be the case when the respective products (e.g., industrial alcohol, medicinal alcohol, or perfume) are taxed at a markedly lower rate than are usual alcoholic beverages .
•Alcohol produced reception or artisanal (either legally or illegally).
•Alcohol that’s procured from any illegal source (smuggled alcohol or industrially produced illegal alcohol).
•Alcohol that’s purchased at duty-free outlets or abroad, and consumed not within the jurisdiction where it’s recorded.
Recorded consumption are often measured via sales and taxation, or via production, export, and import, and plenty of national governments regularly monitor this a part of alcohol per capita consumption , . Harder to get data are required to estimate and monitor unrecorded consumption at the country level. Only some countries, like Sweden, have a daily national monitoring system of unrecorded consumption, which in Sweden had been in situ for over a decade .
Most other countries don’t monitor unrecorded consumption, and only occasionally there had been specific efforts to estimate the impact of unrecorded consumption in research studies (for an outline of studies ; for a scientific effort by the EU to investigate unrecorded consumption in several countries ). Thus, for systematic monitoring of unrecorded consumption within the WHO efforts related to the worldwide strategy to scale back the harmful use of alcohol  and therefore the Global Monitoring Framework for non-communicable diseases , a distinct methodology had to be adopted. With in the next sections, this effort are going to be described intimately.
Monitoring unrecorded alcohol consumption within the WHO Global system on Alcohol and Health
The Global system on Alcohol and Health (GISAH) has been developed to be the most important tool for assessing and monitoring the health situation and trends associated with alcohol consumption, alcohol-related harm, and policy responses in countries (http://www.who.int/gho/alcohol/en/). the most mechanism for gathering data is regular surveys (i.e., iterations of the WHO Global Survey on Alcohol and Health) to any or all WHO membership countries, where information on alcohol exposure (surveys, recorded per capita consumption as derived from regular statistics), alcohol-attributable harm and alcohol policy are collected . This information is complemented by regular systematic searches of the literature, and statistical analyses to estimate missing data (see methods section of the world Status Report , also ), and to triangulate different data sources , , .
The last (2012) also as previous iterations of the WHO Global Survey on Alcohol and Health had questions on unrecorded consumption, but not surprisingly given the issues described above, not many countries indicated to possess data on unrecorded consumption. The systematic search of other sources for unrecorded consumption yielded several hundred publications mostly about chemical composition and health consequences , but not enough data to reliably estimate volume of unrecorded consumption all told member states, even with the foremost sophisticated methodologies for missing value imputation. Thus, it had been decided to supplement the worldwide Survey with a particular study on unrecorded consumption using the Nominal Group Technique , , a variant of Delphi methods , . Delphi methodology denotes a bunch of techniques, originally developed as a scientific, interactive forecasting method, which relies on a panel of experts. it’s now more widely used for any estimates where no algorithmic answers are possible. The experts answer questionnaires in two or more rounds. After each round, an anonymous summary of the experts’ judgments from the previous round is distributed back to the experts additionally because the reasons they provided for his or her judgments. Within the Nominal Group Technique variant, no interaction is feasible between the members of the expert panel. For round one in every of the estimation of unrecorded consumption, the experts were thus supplied with the prior estimate of unrecorded consumption of their country, both in absolute terms (liter pure alcohol adult per capita), and as a proportion of the general alcohol consumed (as an example see the question 1 within the attached questionnaire for Poland within the Appendix). They’re then asked for his or her best estimate, and also the basis for this estimate, and their subjective confidence within the judgment. After providing a summary of round one and therefore the reasons, the identical experts are then asked again for his or her final judgment. The answers of the round two are then averaged as best estimate for the respective country, unless there are better empirical data available. the subsequent countries were included into the study on unrecorded consumption, because they either had high absolute levels of unrecorded consumption, or the proportion of unrecorded consumption was high, or there had been some controversy about the degree of unrecorded consumption within the past: Angola, Azerbaijan, Belarus, Bolivia, Brazil, state, Cambodia, Cameroon, Chad, China, Columbia, Cote d’Ivoire, Democratic Republic of Congo, Ecuador, Estonia, Ethiopia, Ghana, Guatemala, India, Italy, Kenya, Latvia, Lithuania, Mexico, Nepal, Pakistan, Peru, Philippines, Poland, Republic of Korea, Republic of Moldova, Romania, Russia, Rwanda, African country, Sweden, Thailand, Turkey, Ukraine, United Republic of Tanzania, Uzbekistan, Viet Nam, Zambia.
Overall, the experts invited to participate were selected from the literature, reviewed by the Canadian Centre for Addiction and psychological state, and from the respective Ministries of Health (via World Health Organization). It clothed that there was way more expertise on unrecorded consumption than may be found within the published literature.
Estimates supported the subsequent categories were most prominently mentioned:
•Unpublished efforts to estimate unrecorded consumption via proxies . This method has historically been employed in Russia with sugar because the proxy , , . This method has since been abandoned, for using statistic of closely alcohol-related disease categories (like alcohol psychosis or poisoning in Russia) to estimate total consumption and so subtracted recorded consumption to urge unrecorded , . The latter method enables to incorporate not only illegal spirits (“samogon”) supported sugar, but all sorts of alcohol like surrogate alcohol, which plays a crucial role in Russia , . within the example of Poland, one in all the participants presented calculations supported first hospitalizations because of alcoholic psychoses, a technique which had been used historically to indirectly estimate unrecorded consumption in Poland , .
•Information from customs and police on the illegal a part of unrecorded consumption often supported seizures (as the instance of 1 country see the subsequent of the UK: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/293883/03679_Tackling_alcohol_smuggling_Q1_v1_accessible.pdf). When customs officers are invited to the rounds using Nominal Group Technique or other Delphi techniques, they often provided internal data on confiscated alcohol (e.g., counterfeit, for a general review see ; for alcohol counterfeits in Thailand see ). However, as not all unrecorded alcohol is unlawful, and because the seized amount of alcohol may represent only the tip of the iceberg and depend plenty on customs/police activity, it’s very hard to evaluate on the extent of unrecorded consumption from customs and police data alone. While it’s hard to generalize, many estimates supported these data seem to underestimate verity level of unrecorded consumption, either because other sources of unrecorded or because the proportion undetected illegal alcohol are underestimated.
•Small scale regional surveys, or one-time national surveys on unrecorded consumption, which had not been published (grey literature which can include unpublished STEP wise surveys; e.g.www.drugfightmalawi.com/docs/taskforcefinal.doc).
While each of those sources has strengths and weaknesses, the Nominal Group process, while systematically collecting different estimates and allowing local experts to evaluate on their relative strength, will allow a more informed estimate on unrecorded alcohol consumption. However, while the Nominal Group Technique certainly has important strengths in summarizing existing evidence , , , , it relies crucially on selecting the foremost knowledgeable experts in each country to the table, and its results can only be nearly as good because the implicit knowledge bases. during a situation where there’s no knowledge domain in the slightest degree, combining different sources won’t cause an honest estimate.
Shifting the monitoring effort of unrecorded alcohol consumption to the national level
While monitoring of unrecorded consumption on the international level in important, especially in light of the WHO global strategy for alcohol  and of the obligations of the worldwide monitoring framework for non-communicable diseases , monitoring on the national level would be even more important. As indicated above there are in principles four ways to observe unrecorded consumption:
•Using the indirect method described above . While this method is kind of inexpensive once established and may depend upon routine data, the matter could also be that the association between the indirect indicator and therefore the alcohol consumption may change. As mentioned above, the indirect estimation of unrecorded consumption in Russia via sugar, cannot be used, because it wouldn’t cover “samogon” from other sources than sugar, surrogate alcohol or unrecorded wine, which all play a task in today’s alcohol consumption in Russia , , . Also, when drinking patterns change as is also true for Russia , , associations which were supported patterns of drinking like binge and acute outcomes (like alcohol poisoning) may change in strength thus affecting the estimates and therefore the conclusions. Thus, any monitoring effect supported indirect indicators must be re-validated from time to time.
•Direct monitoring via general population surveys is another tool possibility. one among the pre-requisites for this method is that unrecorded must be sufficiently prevalent within the general population prefer it is in Sweden or other Nordic countries, where the most source of unrecorded alcohol is cross-border shopping ,  (for an outline of an ongoing survey based monitoring system in Sweden for unrecorded alcohol consumption see ). For several other countries, where unrecorded consumption is principally prevalent in marginalized and/or institutionalized populations (for example ), this technique works less well, as many such populations don’t seem to be a part of the sampling frame . However, including questions about unrecorded consumption into general monitoring surveys is effective as a primary approximation (e.g., the WHO STEP wise approach to Surveillance survey; http://www.who.int/chp/steps/en/) while this technique is also underestimating verity level. After all, the overwhelming majority of WHO membership nations did indicate little or no knowledge about unrecorded consumption.
•The general population surveys can be augmented with surveys of specialized populations like institutionalized populations. As indicated above, unrecorded alcohol, because it is commonly considerably cheaper than recorded alcohol, is over-proportionally consumed by marginalized folks that drink large quantities of alcohol, and such people may be found within the formal specialized alcohol treatment systems or in social institutions with high proportion of alcohol-dependent people like – looking on the country – halfway houses, shelters or hostels , , .
•Single studies to do to determine all types of unrecorded consumption in an exceedingly country, which after all should be tailored to local circumstances. for instance, so-called “she beens”, alcohol outlets which partly sell unrecorded alcohol and have also been implicated within the relationship between alcohol and communicable disease transmission, may well be used as a part of a research design to estimate unrecorded alcohol in South Africa and surrounding countries , .
Overall, given the extent of the unrecorded consumption and therefore the impact of alcohol on health collectively of the foremost important risk factors for global mortality and burden of disease , monitoring and surveillance efforts for unrecorded alcohol consumption should be initiated or increased also at the national level, and efforts should be made to scale back consumption of unrecorded alcohol (for specific alcohol policy measures see , ). This conclusion may be drawn regardless of the question, whether unrecorded alcohol has an impression on health over and above the effect of ethanol itself (,  for overviews).
1).D.W. Lachenmeier, G. Gmel, J. Rehm Unrecorded alcohol consumption
P. Boyle, P. Boffetta, A.B. Lowenfels, H. Burns, O. Brawley, W. Zatonski, et al. (Eds.), Alcohol: science, policy, and public health, Oxford University Press, Oxford, UK (2013), pp. 132-142,Google Scholar.
2).J. Rehm, J. Klotsche, J. Patra,Comparative quantification of alcohol exposure as a risk factor for the worldwide burden of disease Int J Methods Psychiatr Res, 16 (2007), pp. 66-76,Google Scholar.
3).World Health Organization ,Global status report on alcohol and health
World Health Organization, Geneva, Switzerland (2014)
4).J. Rehm, S. Kailasapillai, E. Larsen, M.X. Rehm, A.V. Samokhvalov, K.D. Shield, et al. A systematic review of the epidemiology of unrecorded alcohol consumption and therefore the chemical composition of unrecorded inebriation, 109 (2014), pp. 880-893
5).R. Room, D. Jernigan, B.H. Carlini, G. Gmel, O. Gureje, K. Mäkelä, et al.
6).J. Shao, P.C. Ivanov, B. Podobnik, H.E. Stanley
Quantitative relations between corruption and economic factors Eur Phys J B, 56 (2007), pp. 157-166 ,Google Scholar.
7).World Health Organization,Global strategy to cut back the harmful use of alcohol World Health Organization, Geneva, Switzerland (2010)
8).J. Rehm, R. Room, M. Monteiro, G. Gmel, K. Graham, N. Rehn, et al.
Alcohol as a risk factor for the worldwide burden of disease Eur Addict Res, 9 (2003), pp. 157-164,Google Scholar.
9).J. Rehm, C. Mathers, S. Popova, M. Thavorncharoensap, Y. Teerawattananon, J. Patra Global burden of disease and injury and economic cost because of alcohol use and alcohol use disorders Lancet, 373 (2009), pp. 2223-2233.Google Scholar.
10).D.W. Lachenmeier, J. Rehm, G. Gmel,Surrogate alcohol: what can we know and where will we go? Alcohol Clin Exp Res, 31 (2007), pp. 1613-1624
11).A. Gil, O. Polikina, N. Koroleva, M. McKee, S. Tomkins, D.A. Leon
Availability and characteristics of non-beverage alcohols sold in 17 Russian cities in 2007 Alcohol Clin Exp Res, 33 (2009), pp. 79-85,Google Scholar.
12).V. Poznyak, A. Fleischmann, D. Rekve, M. Rylett, J. Rehm, G. Gmel
The World Health Organization’s global monitoring system on alcohol and health
Alcohol Res, 35 (2013), pp. 244-249,Google Scholar.
13).M. Ramstedt, Change and stability? Nordic Stud Alcohol Drugs, 27 (2010), pp. 409-423,Google Scholar.
14).H. Leifman,Estimations of unrecorded alcohol consumption levels and trends in 14 European countries Nordisk Alkohol- Narkotikatidskrift, 18 (2001), pp. 54-70,Google Scholar.
15).World Health Organization, Draft comprehensive global monitoring framework and targets for the prevention and control of non communicable diseases World Health Organization, Geneva, Switzerland (2013) Google Scholar.
16).K.D. Shield, C. Parry, J. Rehm, Chronic diseases and conditions associated with alcohol use Alcohol Res, 35 (2013), pp. 155-171,Google Scholar.
17).J. Rehm, T. Kehoe, G. Gmel, F. Stinson, B. Grant, G. Gmel
Statistical modeling of the degree of alcohol exposure for epidemiological studies of population health: the instance of the US Popul Health Metrics, 8 (2010), p. 3
18).T. Kehoe, G. Gmel Jr., K. Shield, G. Gmel Sr., J. Rehm
Determining the simplest population-level alcohol consumption model and its impact on estimates of alcohol-attributable harms Popul Health Metrics, 10 (2012), p. 6,Google Scholar.
19).A.L. Delbecq, A.H. Van de Ven, D.H. Gustafson,Group techniques for program planning: a guide to nominal group and Delphi processes, Scott Foreman and Company, Glenview, IL (1975),Google Scholar.
20).A.H. Van de Ven, A.L. Delbecq,The effectiveness of nominal, Delphi, and interacting group higher cognitive process processes Acad Manage J, 17 (1974), pp. 605-621
21).H.A. Linstone, M. Turoff,Delphi method: techniques and applications Addison-Wesley, Boston (1975),Google Scholar.
22).J. Rehm, V. Gadenne,Intuitive predictions and professional forecasts. Cognitive processes and social consequences Pergamon Press, Oxford (1990), Google Scholar.
23).Y.E. Razvodovsky, Unrecorded alcohol consumption: quantitative methods of estimation,Alcoholism, 46 (2010), pp. 15-24,Google Scholar.
24).V. Treml ,Alcohol within the USSR: a statistical study Duke University Press, Durham, NC (1982),Google Scholar.