Cigarette smoking is recognised as a major risk factor for the development of rheumatoid factor (RF) and the subsequent development of rheumatoid arthritis (RA). In one subset grouping of RA patients with anti-citrullinated antibody-positive arthritis, a combination of the environmental risk factor of smoking and a genetic susceptibility of Human Leukocyte Antigen – DR isotype (HLA – DR) shared epitope showed a significantly increased relative risk for developing RA. Among cigarette smokers, persons with a single copy of the risk gene showed a relative risk of 6.5 and those with double copies a relative risk of 21.0.^1,2

Once RA is established, cigarette smokers have increased joint inflammation with higher swollen joint counts and elevated inflammatory markers, radiological progression, a higher incidence of extra-articular manifestations, including nodules and interstitial lung disease, together with comorbidities such as an increased rate of infections, cardiovascular events and osteoporosis.³ Active smoking can result in a reduced response to treatment, in particular methotrexate and tumour necrosis factor inhibitors, with variable outcomes in current smokers on disease control, joint destruction, inflammatory markers and pain control.^3,4,5,6

In South Africa (SA), the prevalence of smoking has shown a steady decline over the years as a result of many governmental health intervention strategies. This is similar to the trend in other low- to middle-income countries.⁷ The highest prevalence of smoking in SA is in the Western Cape and Northern Cape, and this has contributed to a high burden of both communicable and non-communicable diseases.^8,9

The prevalence of smoking among RA patients in SA, including knowledge about smoking and joint disease and patient attitudes to quitting, is unknown. In light of the multifaceted negative impact smoking has on RA patients, documenting and treating smoking habits and nicotine dependence together with comorbidities in this population plays an important role in long-term disease management.¹⁰ We undertook this cross-sectional study of RA patients attending an academic state-sector hospital with the aim of gaining insight into the prevalence of cigarette smoking, as well as improving our understanding of patient beliefs and readiness to quit. Our primary objective was to describe the prevalence of smoking among RA patients in an outpatient clinic at a tertiary-level public hospital in SA. Our secondary objectives were to: (1) document RA disease control, comorbidities and extra-articular disease in these patients compared to non-smokers or ex-smokers; (2) explore RA patients’ knowledge about smoking and RA disease; and (3) measure nicotine dependence and attitudes to smoking and quitting. Provisional results of this research have been published as a congress abstract.¹¹

Of 632 participants (536 were female), the mean (standard deviation [s.d.]) age and disease duration were 55.4 (13.0) and 10.1 (9.3) years, respectively (Appendix 1 Table 1-A1). Over a third (n = 218, 34.5%) self-identified as current smokers, with 89 (14.1%) ex-smokers. A poor SES was noted in 64.0% of participants, and the group overall had poor anxiety, depression, fatigue, pain severity and pain interference scores, with moderate to severe functional disability. Most patients exhibited low disease activity (48.9%), but the mean patient global score was moderate. Comparing smokers to non-smokers, smokers were more likely to be male and have a significantly lower body mass index. Smokers were more likely to be RF positive and were more likely to have rheumatoid nodules, although this was not statistically significant. There were no differences in age of onset, level of education and family history.

Comparing the disease activity and PROMs of smokers and non-smokers, smokers showed significantly worse anxiety scores, and more impairment on the EQ-5D domains of usual activities, anxiety and depression, with no statistical difference between the two groups for CDAI score and its components, complications of RA, HAQ-DI or HADS depression scores. Smokers were significantly more likely to have chronic obstructive pulmonary disease (COPD) than non-smokers, while other comorbidities did not have significant difference. The main comorbidities were hypertension (58.9% of all patients), diabetes mellitus (27.9%), dyslipidaemia (25.3%) and obesity (7%). Multivariable analyses revealed that smoking was significantly associated only with functional disability as measured by the HAQ-DI (β = –0.26, 95% CI [confidence interval] – 0.48 to –0.05, p = 0.018).

More than half of the patients (56.1%) reported starting smoking at an age of 18 years or younger. The total number of cigarettes smoked per day was reported as less than 10 in 72.8% of individuals. The major reasons participants identified for ongoing smoking were emotional support (45.8%), nicotine craving (30.5%) and pain control (25.2%). Less commonly listed reasons included social factors (22.9%), boredom (20.0%), and coping strategies (17.6%). Many (50.6%) reported that living with RA makes it difficult to quit smoking. Only 48.0% of patients were aware that smoking worsened arthritis, yet 74.4% reported receiving advice to quit smoking from their arthritis doctors. Patients who claimed they smoked for pain control had higher pain severity scores than those who did not feel pain control was a reason they smoked (5.9 (1.6) vs 4.7 (2.0), p < 0.001). Those smoking for emotional support had higher anxiety scores (10.6 (4.9) vs 7.9 (4.5), p < 0.001).

When asked to list the benefits of smoking cessation, the most common responses were: lower risks of heart attack (74.8%), cancer (59.5%) and emphysema (54.0%). Half of the smokers felt that quitting would lead to better control of arthritis (49.7%); others listed financial savings (47.9%), better fitness (37.4%), no longer smelling of smoke (32.5%) and approval from partner or family members (31.2%) as advantages of quitting. Almost half (45.6%) had previously quit for more than 3 months. A small percentage (16.8%) reported being ‘desperate to quit’, and the majority were ambivalent or not interested: ‘maybe give it a try’ (44.2%), ‘not right now’ (32.6%).

The majority of current smokers (67.5%) had mild nicotine dependence as per the Fagerström test, a quarter (24.4%) moderate and only 7.5% severe dependence. The Fagerström score positively correlated with both depression and anxiety scores: r 0.28 (p < 0.005) and r 0.20 (p = 0.02) respectively.

In this cross-sectional study of RA patients attending a tertiary referral centre, over a third of participants were smokers, with a further 14.1% previous smokers. This prevalence of smoking is high in comparison with the national prevalence of 17.6%, but similar to regional numbers, where the smoking prevalence in the Western Cape is 32.9%.⁸ Elsewhere, the prevalence among RA patients shows wide regional variability, ranging from 0.9% (Morocco) to 48% (Austria).^23,24,25

In our study, we found that more male participants smoked, in keeping with national demographic trends, and smokers had a significantly lower body mass index (BMI) with a higher prevalence of COPD.^8,26 Seropositivity, nodules and anxiety were more prevalent in smokers, which has been reported elsewhere.^27,28 We found no significant difference among the PROMs. Many participants in our study reported emotional support and pain control as primary reasons for continuing their smoking habit. These subgroups had higher pain severity and anxiety scores, respectively, reflecting the challenges of living with RA, pain and anxiety and perhaps demonstrating that cigarette smoking is felt to ameliorate these problems.

The association between smoking, pain, anxiety and depression has been well described.^5,29,30 Current tobacco smoking worsens pain, and early stages of nicotine withdrawal amplify pain. Smoking has been associated with other chronic pain syndromes, with worse outcomes reported in pain treatment programmes. Anxiety and depression are more prevalent among smokers relative to non-smokers and in patients experiencing chronic pain. A complex bi-directional relationship ensues with pain and smoking associated with or worsening anxiety and depression, and the negative effects of poor mental health on pain, pain management, smoking habits, and smoking cessation. Improved control of RA disease, better nonpharmacological and pharmacological pain management, together with attention to mental health care, may improve PROMs in conjunction with smoking cessation programmes.^31,32

Smoking is a modifiable cardiovascular risk factor.³³ The comorbidities of non-communicable and lifestyle diseases are prevalent in our RA patients, with the vast majority having two or more comorbidities, including hypertension, dyslipidaemia, diabetes and obesity. Each of these individually contributes towards a significant cardiovascular risk profile in RA patients.^34,35 Smokers have lower BMIs due to appetite suppression, metabolic effects on lipid and glucose pathways and sympathetic activation. However, this does not offset the cardiovascular risks of smoking.^36,37

Of concern is that less than half of the participants believed that smoking worsened their arthritis, although most reported having received smoking cessation advice from their arthritis clinic team. This is an area for improvement and should be ongoing at every clinic visit. Encouragingly, a minority of patients have severe nicotine dependence, with most patients having mild dependence. Similarly, low levels of nicotine dependence but high levels of depression have been reported at our local smoking cessation clinic.³⁸ This, along with the long time to return to a non-smoker risk level, highlights the importance of early and repeated interventions for smoking cessation.³⁹

The high cost of pharmacological therapies available for smoking cessation in low to middle-income countries limits their availability. The majority of smokers in the current study are not interested in or ambivalent about smoking cessation, despite low nicotine dependence. Simple and affordable strategies, including reading material, education, and brief interventions are likely to increase the interest in quitting, with readily available support for those who are actively interested in quitting. Addressing the underlying level of disease control and associated anxiety and depression in parallel with smoking cessation advice is likely to yield the best results, given the high prevalence of smoking.

Several limitations of this study can be highlighted. This is a single-centre study in the Western Cape with high smoking prevalence, which may limit its generalisability to other RA centres. As participation was voluntary, smoking may be under-reported. Smoking status and pack-year history were self-reported; thus, the true prevalence may be under-reported. Other forms of nicotine delivery, for example, chewing tobacco, vaping, or the use of cannabis or other recreational drugs, were not evaluated in this research project, and have been shown to be prevalent.⁴⁰

The prevalence of smoking among RA patients in the Western Cape is high and strongly associated with pain, anxiety and depression. Many smokers with RA were unaware of the direct negative effects of smoking on their joint disease, despite having knowledge of the risk of cardiovascular, cancer and respiratory disease. It is critical to interrupt the cycle of uncontrolled disease with further smoking-induced inflammation. A multidisciplinary approach addressing mental health, joint pain and reduced quality of life along with smoking-specific behavioural modification and nicotine replacement therapy, where needed, is required.