Correlates of caring for the drinkers and others among those harmed by another’s drinking
Abstract
Introduction and Aims. This study identifies the correlates of caring for harmful drinkers and others, and examines how caring for that person impacts on respondents’ well-being and use of services. Design and Methods. The study utilises the data from the 2008 Australian Alcohol Harm to Others Survey (n = 2649), in which 778 respondents reported they were harmed because of the drinking of someone they knew. Respondents were asked about the person they were most adversely affected by and whether they spent time caring for this person because of their drinking. Logistic regression models are developed to examine which factors were associated with the prevalence of caring for others. Results. The study reveals that the respondents who cared for others because of the other’s drinking reported lower quality of life than the respondents who did not have to do this.The results of the logistic regression suggest that respondents were more likely to care for the drinker if the drinker drank more (as the usual quantity of alcohol consumed increased), but less likely to care for the drinker if the drinker drank five or more drinks on more than four days per week. Discussion and Conclusions. The findings of the study suggest that the drinking of family and friends can be a substantial burden for their households, families, friends and others. Policy approaches that reduce the amount of heavy drinking, particularly heavy drinking in a single occasion, are likely to reduce the burden of caring for others because of other’s drinking. [Jiang H, Callinan S, Laslett A-M, Room R. Correlates of caring for the drinkers and others among those harmed by another’s drinking.
Key words: alcohol, caring for others, family, friend, quality of life.
Introduction
The negative effects experienced by family members, households, friends and co-workers because of drinking by others are substantial [1]. Although a number of studies [2–4] have investigated the adverse effects from others’ drinking, there is a lack of research on caring provided because of the drinking of others.
Living with a relative who misuses alcohol and illicit drugs can cause many negative experiences for family and household members, such as strain, violence, poverty and social isolation [5]. A study in the USA found that caregivers of people with mental health and drinking problems were more likely to report worse physical health, including insomnia, headache and weight loss, higher rates of depression and anxiety, and a poorer quality of life than non-caregivers [6].
While there has been little research done specifically on caring for drinkers, objective negative effects on family members from problematic drinkers have been examined by Orford and Dalton [7]; interviewees reported driving their family members home or to other places because of their drinking, that regular drinking made the drinkers unable to do their usual household tasks and meant that they were sometimes ‘lazy’, and that their family members had to spend extra time caring for them or others in these and other ways.
The alcohol-related harm to others study in New Zealand found that 35% of respondents (n = 760) reported that they had to spend extra time taking drinkers somewhere because of their drinking. Further- more, 39% of respondents reported they had to clean up after others’ drinking, and 30% of respondents reported they had to take on extra responsibilities caring for children or others because of someone else’s drinking [2]. Such burdens have also been reported in relation to other conditions. For example, among fami- lies of schizophrenic patients, Hoenig and Hamilton [8] reported objective burdens upon carers—exemplified by financial loss and negative effects on health—and subjective burdens, such as abnormal behaviours or psychological effects.
Studies examining those who require care because of dementia, schizophrenia and cancer have found that most of the care is provided by family members, spouses and friends [9]. Similarly, we may assume that in the case of those who require care because of their drinking, family members, spouses and friends of the drinker may look after them and their dependents. A US and a British study found that families and friends were identified as the primary sources of social support to persons with substance use disorders; they provided direct care as
well as financial support [10,11]. There is little other evidence in the literature on the issue.
Informal caring (and how the caregiver is affected by this caring) will be influenced by the behavioural or cognitive problems of the person needing care and by the characteristics of the caregiver (e.g. age, sex, and social and economic status) and the social context (e.g. finances and social and family support) [12]. Further- more, a number of variables concerning the person needing care (e.g. if the drinker is a child of the respondent) will be important predictors of whether those around the drinker take on the role of caring for the drinker and whether the caregiver is harmed or disadvantaged by the experience.
Another factor that will likely impact on the burden of care is the drinking patterns of the drinker. A drinker is temporarily
incapacitated in several ways while intoxicated and while recovering from intoxication, and incapacities may become more lasting with recurrences over time. This has primarily been studied concerning the relation of drinking pattern to problems for the drinker himself/herself. Herd [13] reported that the frequency of being intoxicated and the drinking pattern were both important predictors of alcohol-related prob- lems and concluded that drinking five or more drinks was significantly associated with alcohol-related prob- lems. A number of studies found that high levels of alcohol-related problems were strongly and positively related to drinker’s frequency of 5+ drinking and binge drinking pattern [14,15]. While an increased frequency of heavy drinking and drinking problems among those in one’s life may be assumed to be important predictors of increased experience of caring for others because of their drinking, this has not been directly assessed.
This study aims to identify which factors are corre- lated with whether the respondent takes on this caring role for the person in their life whose drinking has most adversely affected them in the current year and to examine how caring for that person impacts on the respondent’s quality of life and well-being, and use of services.
Method
Sample
The data used in this study were taken from the Range and Magnitude of Alcohol’s Harm to Others survey. This national survey of the Australian general popula- tion was undertaken in November and December 2008 (see Wilkinson et al. [16] for more details). The survey collected a range of data on the impact of alcohol on people other than the drinker, using computer-assisted telephone interviews. The sample was generally repre- sentative of the national adult population [16]. In total, 2649 interviews of those aged 18 years or older and who spoke English were conducted. A cooperation rate of 49.7% and a response rate of 35.2% were achieved. The study was reviewed and approved by the Victorian Department of Human Services Ethical Review Committee.
Measurement
In the Range and Magnitude of Alcohol’s Harm to Others survey, respondents first listed the number of heavy drinkers known to them by relationship type. Approximately 30% of 2649 respondents (n = 778) reported they were harmed because of the drinking of someone they knew. They were then asked to identify the person ‘whose drinking had most negatively affected them’, that is, the most harmful drinker (MHD) (see [1] for more explanation). Later in the interview, they were asked four questions regarding how often and in what ways they cared for the MHD or others related to the MHD because of the MHD’s drinking. Please note that in these questions, ‘him or her’ or ‘his or her’ refer to the previously identified MHD.
1. How many times in the last 12 months did you have to spend time caring for him or her because of his or her drinking?
2. Did you have to take extra responsibilities caring for children or others because of his or her drinking?
3. Have you had to clean up after him or her because of his or her drinking?
4. Did you have to drive him or her somewhere or pick him or her up because of his or her drinking?
For the current study, responses to these items were aggregated; endorsement of any of the four items was defined as indicating caring.
The relationship of the drinker to the respondent was summarised as: in the household, family member not in the household, partner, friend, co-worker and others. The ‘other’ group includes neighbours, former housemates, acquaintances and other types of relation- ships. Furthermore, some specific variables were included in the analyses for examining the drinking behaviour of MHDs and respondents. The average drinking quantity of the MHD when they drank heavily was summarised in terms of <5, 5–6, 7–10, 11–19 and >19 standard drinks per drinking occasion. Usual drinking quantity patterns of respondents were defined as <5, 5–6, 7–10, >10 standard drinks per drinking occasion. The usual frequency with which MHDs drank 5+ drinks and the respondent’s usual drinking frequency both were classified as <3, 3–4 and >4 days per week.
The health and well-being of all respondents were measured using the Personal Wellbeing Index (PWI) and a health-related status index (EQ-5D), which have both been widely used as the indicators of quality of life in many studies [17–20]. The EQ-5D, a standard and non-disease-specific index, was introduced to measure the health-related quality of life. More details about the scoring method of PWI and EQ-5D are elaborated in the PWI manual [21] and The EuroQol Group [22]. Respondents were also asked about service use attrib- utable to the drinking of others, that is, if they called the police, were admitted to hospital or an emergency department, received other medical treatment or sought help from a counsellor or other health profes- sional, or rang a helpline.
Results
Descriptive statistics
As noted in the Methods section, the sample in the current study is made up of the 778 respondents who stated that they had at least one harmful drinker in their lives. The social and demographic statistics of respond- ents who reported being harmed overall, and then those who did and did not care for others because of the MHD’s drinking are shown in Table 1. Approximately 57% (496 out of 778) of respondents who reported being harmed by an MHD were female; a majority of these among both males and females reported caring for others. Nearly 80% of the respondents among 18–30 years old reported caring for others because of another’s drinking, a higher proportion than in other age groups.
Variations between caring and not caring for others by the type of relationship to the respondent were observed, and the χ2 statistic results for the bivariate analyses suggest that these differences are significant 6 26.4, P < 0.001). High proportions of respond- ents reported they had cared for others where the MHD was a friend (73%), household member (70%) and partner (69%). In contrast, only approximately 54% of those whose MHD was a family member not in the household provided care because of the MHD’s drinking in the last 12 months, while the percentage of respondents who reported provided care where the MHD was a co-worker or in the other category was 58% and 51%, respectively. Demographic predictors of caring The results of bivariate and multivariate logistic regres- sion analyses of caring for others among those who have been affected by others’ drinking are presented in Table 2. The bivariate analysis highlights the likelihood of respondents reporting providing care did not differ by gender. However, it did differ by age, relationship, MHD’s average drinking quantity, MHD’s usual 5+ drinking frequency and respondent’s usual drinking patterns and frequency. Young adults were more likely to have provided care than older age groups. Where a friend was the MHD, care was more likely to be provided than where the MHD was a family member not in the household, a co-worker or with ‘other’ relationship. Compared with MHDs who drank four or less stand- ard drinks per heavy drinking occasion, MHDs who drink 5–6 or 7–10 standard drinks per occasion were approximately 4–5 times (P < 0.05) more likely to be cared for by the respondent, and those who drank 20 or more standard drinks per occasion were approximately seven times (P < 0.001) more likely to be cared for by the respondent. In contrast, if the MHD drank more than four days per week (odds ratio 0.65, P = 0.039), they were approximately two times less likely to be cared for by the respondent than those drinkers who drank occasionally. Overall, these results suggest that the quantity the MHD consumed at heavy drinking occasions is more important than how often they drank when predicting whether or not they were cared for. Indeed, a higher frequency of drinking of five or more drinks of the MHD was a negative predictor of caring for the MHD, children or others. At the same time, if the caregivers drank more often, they were less likely to report caring for others. The multivariate model, with controls for the pres- ence of all other entered variables, indicates that MHD’s drinking quantity, specifically the amount con- sumed when they were drinking heavily, is the main factor predicting the provision of care. An interesting nuance on this finding is that, on the other hand, higher frequency of heavier drinking by the MHD and higher frequency of drinking by the respondent both predicted less caring. Respondents over 50 years old were signifi- cantly less likely to report caring for others compared with younger respondents. Respondents related to but not living with the MHDs were 50% less likely to report caring for ‘others’ compared with friends. Compared with caring for heavy drinking friends, respondents reported no significant difference in caring for all other relationship types, after adjusting for heavy drinking quantity and frequency. Personal and services use outcomes for informal caring Previous analysis of the present data by Laslett et al. [24] indicated that the health and well-being of respondents who reported being harmed by MHDs (PWI = 73.4 and EQ-5D = 81.8) were significantly worse than those who have no MHD in their lives (PWI = 77.7 and EQ-5D = 86.6). How the extra caring responsibilities affected the quality of life of the caregiv- ers was further investigated in this study (Table 3). Among those with an MHD, PWI scores significantly differed between caregivers and non-caregivers ( Pa 0.05), with the non-caregivers showing a sig- nificantly higher well-being status than caregivers. Par- ticularly, respondents who cared for an MHD who was in their household or a friend reported significantly lower personal well-being index (PWI = 69.4 and 72.6) than those who were harmed but did not do any caring (PWI = 75.2). On the EQ-5D measure of health- related quality of life, there was no significant difference between those who had to care for their MHD and those who did not ( Pa 0.05), and differences were also not significant for specific categories of MHD relationship. The prevalence of use of different services by car- egivers and non-caregivers because of drinking by others in the last 12 months is also summarised in Table 3. Overall, one in four informal caregivers had called police and only approximately one in seven reported that they had asked for help from public health services, counsellors, health professionals or help lines. The percentages using each type of service among caregivers whose MHD was a household member or either a family member not in the house- hold or in ‘other’ relationship were significantly higher than for non-caregivers (Pb < 0.05). Discussion This study has presented an overview of the prevalence of caring among those with harmful drinkers in their lives. These extra responsibilities comprise extra caring because of the drinking, not only for the drinker but also potentially for others. The majority of the caregiv- ers were female. The analyses of specific types of rela- tionships and experience of caring for others indicate that it is where the MHD is a household member, or a partner or ex-partner that caring is most likely to occur. The results of bivariate and multivariate models in this study highlight a significant and positive associa- tion between a higher average drinking quantity of the MHD when they are drinking heavily and the burden of caring. This is consistent with previous studies that indicate that the problematic drinker’s drinking pat- terns were positively related to alcohol-related harms [13,14]. However, the drinker’s 5+ drinking frequency was negatively related to the caring. A possible expla- nation is that daily or very frequent heavy drinking behaviour may mean that the risky drinker’s family, friends and households may become less sympathetic and leave the drinker to look after themselves. With- drawing or staying away from a heavy drinker, and gaining some independence from them, is one of the main types of coping style that family members or friends affected by heavy drinkers may adopt [25]. It is also possible that a frequent drinker causes less trouble for a given amount of drinking because of higher toler- ance or because their lifestyle is more adjusted to their drinking. These results may suggest that very heavy but occasional drinking is the primary factor for predicting prevalence of caring for others. Interestingly, the respondents’ own usual drinking frequency pattern was also negatively related to caring, possibly because of the fact that those who consume alcohol more often are not in a position to care for others. The informal caregivers reported significantly lower well-being than the non-caregivers, particularly for those whose care was for an MHD who was a house- hold members or friend. This result may suggest that the care of others comes at a cost. Caregivers were also more likely to ask for help from (and need and utilise) police and health services in the community than the non-caregivers. However, the relatively low rate of police and health service use reported by caregivers suggests that even if caregivers are affected by others’ drinking, they are not commonly likely to utilise police and health services. Whether this is because these ser- vices are inaccessible or not perceived to be appropriate is not known. The positive link between the prevalence of caring and the harmful drinker’s drinking patterns found in this study also suggests that policy approaches that reduce heavy drinking at the population level and indi- vidual risky drinking are likely to reduce the burden of caring for others because of their drinking. Appropriate treatment and intervention strategies can motivate heavy drinkers to moderate their alcohol consumption [26]. Preventive, intervention and treatment strategies for controlling alcohol problems and risky drinking need to understand and take account of the processes of informal control in the drinker’s daily operations with their relatives and friends [27]. In this approach, it may be possible to garner the persuasive powers of the heavy drinker’s partner, households or friends, enabling greater utilisation of public health services. Perhaps, involving these carers in alcohol abuse screening and routine management is an option [28]. There are indi- cations that treatment interventions can work. In a recent longitudinal study, the successful treatment of a family member’s alcoholism helped to reduce the burden on caregivers and improved their quality of life [29]. The 5-Step Method implemented in the UK and Italy can help the problematic drinker’s family to treat and cope with heavy drinking behaviour and reduce alcohol-related harms for family members [30]. Who should provide and how best this assistance might be provided are not clear. For example, the majority of specific alcohol services today still focus on the needs of drinkers themselves. While our study is the first study analysing the preva- lence of informal caring because of drinking by others based on a nationally representative sample, the low response rate is a limitation. The service use may not only have resulted from being harmed by the MHD but also may be the result of being harmed by someone besides the MHD’s drinking (e.g. other harmful drink- ers or strangers). Although caregivers reported using more services than the non-caregivers, the rates of service use by them were still fairly ‘low’, particularly for the respondents whose MHD was a friend.This may be because the caregivers often do not know how to ask for help or refuse to do so because of shame and fear of stigma [31]. Another limitation of the study is the small number of observations for some of the different rela- tionship types between the MHD and the respondent. Further breakdown for the types of relationship would have resulted in low confidence intervals in the estima- tion. Future research that breaks down overall caring into separate components of caring could help to examine whether the correlates differ on experience of different caring forms. The caring for other drinkers (not the MHD) was not measured in the survey. If the MHD was not the person that caregiver spent time caring for, we missed some information on caring. However, this missing information is not necessarily random and should be fairly rare in the study sample. The participants may perceive the term ‘caring’ differ- ently. For example, some participants may think cooking for one person, finishing a drinker’s forgotten jobs and looking after their children are a part of their daily routine and not consider these extra caring responsibilities as a burden of caring because of anot- her’s drinking. Therefore, the measurement of caring for others because of drinking by the MHD could be underestimated. The results of this study reveal the substantial burdens placed on households, relatives and friends caring for others because of their drinking. It focuses attention on their reduced well-being and suggests that services are rarely utilised to meet their own needs. Links between caring and the consumption of those being cared for suggest that policy changes that would reduce consumption, particularly heavy drinking on an occasional basis, NVP-DKY709 would in turn reduce the impact and/or the prevalence of caring for others because of their drinking.