Performance of the WHOQOL-BREF among Norwegian substance use disorder patients

Reducing morbidity and improving the quality of life years spent with or in recovery from a substance use disorder (SUD) is as key to treatment as reducing mortality. In the management of chronic diseases, in which the negative effects of a disease are often reoccurring and long-lasting, treatment cannot be evaluated by a cure, but by sustained improvements in self-assessed functioning and well-being. The multidimensional measure of patient-reported outcomes such as quality of life (QoL) are important outcomes of the treatment of chronic diseases such as SUDs, as it captures the repercussive impacts of disease in many areas of life that may not be identifiable through normal clinical practice [1, 2]. QoL is of further importance in SUD treatment because improving QoL may itself support treatment retention and abstinence after leaving treatment [3‐5].

Reduced substance use is often assumed to automatically improve QoL. However, mixed evidence for this hypothesis among various SUD groups exists [6‐9]. People may even instigate SUD treatment less to reduce their substance use than to improve their QoL in general [4, 10]. Qualitative research has shown people with SUDs consider social factors such as enough relationships and contact, a supportive network, and social inclusion as necessary components to good QoL [11‐13]. Indeed, these social settings may influence the development and trajectory of SUDs in general to a far greater extent than other chronic diseases, as substance use in a person’s network impacts their own access to both substances and, later, to treatment [14‐16].

Among non-SUD populations, several social determinants have been identified: Pinquart and Sorensen’s [17] meta-analysis found that the quality of social relationships among the elderly was the second most important factor in their QoL, following independence. In a cross-sectional study of people with severe mental illness, higher QoL correlated to greater social support, lower internalized and societal stigma, and lower symptom distress [18], and in longitudinal studies of people with various mental illnesses, improvements in social integration along with symptom reductions have predicted QoL improvements over time [19‐21]. Sociodemographic factors, age, and gender have not been consistent predictors in these studies.

Overall QoL was reported in the aforementioned studies through single or aggregate scores [17‐19, 21]. Instruments can also measure QoL multidimensionally, in which separate domains of respondents’ lives are queried and represented through separate domain scores. The gold standard in generic quality of life instruments, the World Health Organization’s brief quality of life tool (WHOQOL-BREF), measures four domains of QoL: physical health, psychological health, social relationships, and environment QoL. Strong psychometric properties have been reported internationally [22] as well as among the general populations of Nordic countries [23‐25]. Normative values for this generic instrument allow for the comparison between healthy and clinical groups, such as those with SUD, and to this end, the WHOQOL-BREF has also been validated among opioid maintenance treatment patients [26‐28], a substance-using veteran population [29], and most recently among an exclusively alcohol-abusing sample [30].

Two reviews have concluded that most QoL instruments administered to people with SUD under-prioritize social domains in favour of health domains, and most research on QoL determinants continues to focus on substance use and mental health [8, 31]. Two disease-specific QoL tools – the Injecting Drug Users Quality of Life Scale [32] and the Opioid Substitution Treatment Quality of Life scale [13] – both include a broader focus on social domains, but disease-specific QoL tools cannot provide QoL scores that can be compared to other populations; a recent review discusses further merits of generic measures among opioid users [33].

The WHOQOL-BREF may be a preferable generic instrument among SUD populations, given that one of its strengths is the inclusion of a social QoL domain. QoL instruments such as the WHOQOL-BREF, if they are to allow for the subjective evaluation of the effects of SUD and treatment, must also be feasible and perceived as relevant to respondents. The WHOQOL-BREF has seldom been reported for Norwegian substance users, with a small pilot study being one exception [34]. We wished to explore the acceptability of the WHOQOL-BREF to SUD patients in Norway. We analysed the WHOQOL-BREF’s item responses, responsiveness, scaling qualities, and explored the relationships of various health, substance, and social variables to each domain.

In the first such analysis of the WHOQOL-BREF in Norway, lacking regular (reported) social contact had a strong negative effect on numerous dimensions of QoL of 107 participants in a substance use disorder treatment study. Along with depression, social factors were more important than direct effects of substance use or of severity indicators. The WHOQOL-BREF exhibited satisfactory scaling qualities and user acceptability and is therefore overall considered a useful tool among this population, but its social relationships domain needs further investigation. We recommend that the social lives of people with SUD receive more attention in future QoL instrument development and in treatment.

The poor QoL reported by this sample is in line with a larger body of research on SUD populations [7, 43, 44]. Of novelty in this analysis is our exploration of social contributors to QoL. This sample reported concerningly low levels of social contact and engagement, with 18% reporting being isolated, 57% living alone, 77% without a partner, and 71% disconnected from both the labour market and educational system. A qualitative study of OMT patients in Belgium reported a similar paucity of social contact, and patients linked their lack of daily activities to feeling alone and without emotional support [45]. The effects of such social disengagement in our sample extended beyond the social relationships QoL domain, and low social contact had negative interaction effects in every other domain. Social isolation appears so detrimental to well-being that it overrode the positive effect of exercising in the environment QoL domain. The environment domain may also indicate social inclusion, as it measures aspects such as access to social care and participation in leisure activities. This sample reported slightly lower environment QoL than social relationships QoL.

Other negative health-related variables such as smoking and being dissatisfied with one’s weight, in tandem with social isolation, reduced the physical health QoL domain. Smoking and weight dissatisfaction are common among people in SUD treatment [46‐48], which is particularly concerning given evidence that smoking cessation is associated with higher rates of long-term abstinence from other substances [49], yet is not standardly integrated into treatment programs [50]. More than half of the sample reported symptoms of clinical depression, which was a component of poorer QoL in every domain, in alignment with extensive international [7, 8, 51], Scandinavian [46], and Norwegian research [52] reporting mental health as a factor in QoL among SUD populations. We found no correlation between active substance use or treatment variables to the physical health, psychological health or social relationships domains of QoL, contributing to existing mixed evidence of these relationships among people with SUD [6‐9, 53, 54]. The environment QoL domain was most responsive to substance use variables in bivariate analyses and to a lesser extent in the adjusted model, and the validity of this domain may be higher for people in active substance-using phases than for those in other populations.

Many of the components of poor QoL seen in this analysis can be addressed in treatment. When depression is addressed, symptoms can improve significantly and their corresponding treatment barriers alleviated [55]. Similarly, lifestyle interventions among SUD populations have been successfully implemented to reduce smoking and encourage other health-promoting behaviours such as healthy eating, sleep hygiene, and exercise [46, 56‐58]. Social contact itself can support healthy behavioural change, as reported in a recent qualitative analysis of alcohol treatment patients who dropped out of an exercise intervention [59]. These results suggest that in addition to mental health and exercise, specific interventions aiming at improving social contact should also be a focus during treatment.

Social isolation was operationalized through answering the question, “with whom do you spend most of your free time?” with “mostly alone”. Patients may have responded differently had they been asked directly if they were isolated. Being mostly alone could be a temporary situation resulting from a recent change in social situation, such as loss of a job or a partner, or following relocation. It could also be an intentional technique utilized by those who would otherwise only have substance-using people around them, as described in previous qualitative studies [60‐62] At the same time, people in recovery have also reported loneliness as an instigator for relapse [63, 64]. As Mau et al. [65] have recently discussed, while the agency of people with SUD in negotiating their own loneliness and isolation should be acknowledged, “loneliness should not be clinically accepted as a requirement or side-effect of recovery in the long-term” (p.12).

The WHOQOL-BREF performed well as a whole, with the exception of the social relationships domain. Expected social variables such as isolation or relationship status did not explain observed variance in this domain, and in fact, our model did not adequately explain variance. Most likely, the three questions that comprise this domain – how satisfied are you with your personal relationships, support from friends, and sex life? – are simply too few, and should be supplemented with items from the larger WHOQOL-100 instrument. The negative interaction effects of low social contact on almost all domains, along with the low internal consistency of the social relationships domain, underscore the need for QoL tools to capture the social lives of people with SUD and likely other marginalised groups, but the importance of social factors is belied by the size of this domain in the WHOQOL-BREF. This may require the development of a new generic QoL tool with a greater focus on social QoL.

The heterogeneity of this sample – including users of different substances and having entered different treatment modalities – is a strength of this study, as the SUD population itself is diverse. Nevertheless, this sample was relatively small, prohibiting stratified analyses such as by gender. No gender differences in QoL were found in previous analyses from the larger study from which this sample was drawn [42, 66], but there may be a gender aspect to social networks and contact, and it is worth exploring this further. Another limitation is that the larger study did not utilize a random sampling design, limiting generalizability of the study’s and of this sample’s characteristics to the Norwegian patient population. While this analysis is therefore less capable of providing specific prevalence rates, the correlations found between QoL, exercise, and social contact seem stable, as they were also found in analysis of the larger study at baseline [42] and follow-up [66].

The WHOQOL-BREF is a short, valid measure of QoL among a SUD treatment population and exhibited acceptable scaling properties and item discrimination. More attention to the social relationships domain is required, however, and better information may be collected through additional questions/developments. The negative effect of low social contact with other variables was seen in every domain of QoL, and treatment providers should monitor the contact patients report when they enter treatment, during treatment, and at discharge. Addressing the quality and functionality of the social contact and support that patients have may be an important part of intake and future treatment processes, in order to best support patients’ engagement with people supportive to their recovery. These findings may also be relevant for other patient groups with high rates of isolation, and a focus on increasing social contact may be an important element to add to the health care of marginalized populations.