Socioeconomic deprivation independently predicts symptomatic painful diabetic neuropathy in people with type 2 diabetes

Peripheral neuropathy is a common microvascular complication of diabetes, the incidence of which increases with age and with poor glycaemic control. Diabetic neuropathy is the most common form of neuropathy in the western world (Young et al, 1993; Pop-Busui et al, 2009). Diabetic peripheral neuropathy increases the risk of recurrent lower extremity infections and ulceration, and is a major contributor to non-traumatic lower extremity amputations (implicated in more than 60% of cases; Tesfaye et al, 1996; Abbott et al, 2011).

Painful diabetic peripheral neuropathy is a debilitating condition with adverse psychological and occupational implications. A previous study in northwest England showed that an estimated 35% of the population with type 2 diabetes had painful symptoms, with the prevalence of painful diabetic peripheral neuropathy being 21% (Tesfaye et al, 1996).

We have previously demonstrated an independent relationship between social disadvantage and prescriptions for painful diabetic neuropathy in people with type 1 diabetes (Anderson et al, 2014). In the current community-based study, we investigated the prevalence of symptomatic diabetic neuropathy in relation to indices of socioeconomic deprivation in people with type 2 diabetes from the county of Cheshire. We also analysed prescription trends with respect to pharmacotherapy for neuropathy pain relief in this population.

Methods
We examined pseudo-anonymised records in 15387 people with type 2 diabetes attending general practices in the catchment area of our service in Central and Eastern Cheshire Primary Care Trust. We included all individuals with type 2 diabetes who were on our register in early 2012. Overall, 6770 participants (44.1%) were women and the mean age was 67 years (range, 16–101 years).

We used the Townsend deprivation index (Townsend et al, 1987), based on participants’ postcode, to assess socioeconomic deprivation. This index was devised to provide a material measure of deprivation and disadvantage in a population, and scores were derived from census variables taken originally from the 1991 UK Census. More positive Townsend scores are associated with geographical areas with high deprivation. Negative values are associated with relative affluence.

Results
Symptomatic neuropathic pain of a degree requiring pharmacological treatment was present in 3266 participants (21.2%), of whom 973 (6.3% of the total cohort) had a formal diagnosis of peripheral neuropathy recorded. The agents prescribed are shown in Table 1 and Figure 1. Specifically, 2162 participants were prescribed one agent, 756 received two, 263 received three and 85 received four or more.

Of the 15387 individuals with type 2 diabetes, 2748 (17.9%) had a diagnosis of either depression or mixed depression and anxiety disorder. Of the 3266 people receiving pharmacological treatment for painful diabetic neuropathy, 1000 (30.6%) had one of these diagnoses.

Townsend index scores ranged from –6 to +8, with higher scores relating to increased social disadvantage. When participants were categorised according to whether they received treatment for neuropathic pain or not, there were significant differences between the groups in Townsend index scores, with a greater proportion of those receiving treatment having a score of ≥1 (30.6% vs 22.8%; chi-squared=83.9; P<0.0001).

In univariate analyses, each unit increase in Townsend index score was associated with a 6% increased risk of having painful neuropathy requiring pharmacological treatment. Figure 2 displays participants’ probability of having neuropathic pain requiring treatment in relation to Townsend index score.

These findings were supported by multivariate analyses, in which, again, each unit increment in Townsend score was associated with a 6% increased likelihood of having painful neuropathy. In these analyses, the probability of having painful peripheral neuropathy requiring treatment increased by 5% for each 5-year increase in age, and by 3% for each unit increase in BMI. There was a 3% reduction in risk for each 5-unit increase in estimated glomerular filtration rate (indicating better renal function), and a 3% reduction for each 5-mmHg increase in systolic blood pressure. There was no relationship with HbA_1c or circulating cholesterol levels. Exclusion of the 1000 participants with depression or mixed anxiety/depressive disorder did not change the risk estimates in the multivariate model.

Discussion
We have found a significant independent association of social disadvantage with the risk of symptomatic diabetic peripheral neuropathy in people with type 2 diabetes. This was independent of glycaemic control, suggesting that lifestyle factors and societal disadvantage may play a role in the aetiology of neuropathic pain independent of their influence on metabolic control.

The results reflect our previous findings that a higher level of socioeconomic deprivation may independently predispose to severe neuropathic pain in people with type 1 diabetes (Anderson et al, 2014). At 21%, the overall prevalence of painful diabetic neuropathy in our cohort was comparable to that found in an earlier community-based study in northwest England (Tesfaye et al, 1996).

The existence of an inverse care law – in that people in greatest need of healthcare are the least likely to receive it – is important to consider in the context of our data. Healthcare providers’ willingness and ability to provide comprehensive complication reviews for people with diabetes, and their knowledge of newer trials supporting the benefits of intensive management, are likely to be worse in deprived areas (Franks et al, 2003). In publicly funded healthcare systems such as the NHS, inequities like these can be addressed.

We also studied drug prescription patterns for painful neuropathy in our cohort. Amitriptyline was the most common agent prescribed to treat neuropathic pain. Current guidelines issued by NICE (2013) for the treatment of neuropathy in a non-specialist setting, usually primary care, recommend any one of amitriptyline, duloxetine, gabapentin or pregabalin as first-line therapy, with a choice of one of the remaining three agents in case of ineffectiveness or intolerance. Capsaicin cream is permitted for people with localised neuropathy who are intolerant to or wish to avoid oral medications. There was a high rate of tramadol prescribing outwith NICE guidance for long-term management of neuropathic pain.

Study limitations

There were a number of limitations to our study, mainly related to the fact that we relied on pseudo-anonymised clinical records from family practitioners. This was, therefore, not a systematic study in which all participants had an identical assessment. It is also possible that, in a small number of cases, the aetiology of the pain was neuropathy unrelated to diabetes. Furthermore, the study was retrospective in relation to collection of data.

Conclusion
In conclusion, a higher level of socioeconomic deprivation appears to predispose to severe neuropathic pain in people with type 2 diabetes. In a publicly funded healthcare system like the NHS, this would suggest a need for targeted allocation of healthcare resources in areas of relative disadvantage, as well as a focus on preventative strategies through healthcare education and focussed diabetes management.