These studies were almost all database retrieval studies. A wide variety of case identification methods was used, often involving a number of different steps to verify the cases identified. For example, researchers in the study by Koopman et al.  used a 2-step process of case ascertainment: first, the researchers used the International Association for the Study of Pain (IASP) case definition criteria to search the database for different facial pain conditions associated with neuropathic pain and classed as either probable, possible, or no neuropathic pain; all probable cases were reviewed by a second medical doctor and discrepancies were arbitrated by a pain specialist. Secondly, additional information from the patients' medical records was sought to verify all possible cases.
Prevalence rates of PHN were reported in 2 included studies [9,26]. They ranged from 0.09% per annum and 0.7 per 1000 (Table 4). The study by Di Luzio Paparatti et al.  contained missing data. They reported that 408 of the 95,508 patients in their population had herpes zoster (HZ); however, only 322 of those were screened for complications. Of those screened, 21.4% had PHN. For the purposes of this analysis, we assumed that the rate for those screened was the same as for those not screened. The prevalence therefore was calculated by the equation [(408*21.4)/98,508].
The incidence rates of PHN were reported in 9 studies [6,10,18,19,21,24,26,30,32]. Note that some articles calculated incidence rates in percent of population per year [6,30,32], whereas others reported per 100,000 person–years [10,18,19,21,24], and one reported number per population per year . For this reason, direct comparisons could not be made among all studies. For those that were comparable, rates ranged from 3.9/100,000 person–years to 42.0/100,000 person–years (Table 3).
The study by Opstelten et al.  (2002) was calculated from the data provided over a 5-year period, and the number in the total study population varied from year to year. Of the 837 individuals with HZ, 2.6% had persistent pain after 3 months. As the overall population was reported to be approximately 49,000, the rate was calculated by the equation: [(2.6*837)/49,000)] over 5 years; dividing the answer by 5 found the incidence rate per population per year. A similar method was used in the study by Pierik et al.  to calculate the incidence.
The study by Choo et al.  (1997) had reported a prevalence of PHN among those who had experienced HZ in a general population said to number 250,000. In fact, it was incidence that they calculated. There were 37 cases of PHN identified >60 days after HZ infection. The rate of PHN in the general population was recalculated using the equation [37/250,000] and divided over 2 years.
Incidence rates of TN were reported in 7 studies [10,18,19,22,24,26,32]. Comparable rates ranged from 12.6/100,000 person–years to 28.9/100,000 person–years (Table 3).
The population prevalence of PDPN was calculated at 0.8%, representing 26.4% of those with type 2 diabetes who responded positively to screening questions and attended clinical assessment (Table 4). Incidence rates of PDPN were reported in 3 studies [10,18,19]. Rates ranged from 15.3/100,000 person–years to 72.3/100,000 person–years (Table 3). PHN, TN, and PDPN had the highest incidence rates, compared to GN and PLP.
3.4. Funding of included studies
We identified 3 main sources of funding for the research. Some studies were funded by government, charity, or nonprofit organisations [8,15,21–23,26,30,38]; others had educational grants from industry but no authorship [4,6,7,10,24,35,42]; and the remainder were conducted and/or supported by pharmaceutical industry and had declared authorship [9,13,14,18,19,32]. There were no apparent systematic differences in study results associated with the source of funding.
This is the first comprehensive systematic literature review of epidemiological studies of neuropathic pain in the general population. Rather than a traditional systematic review of randomised controlled trials, this review is based on observational studies within the general population using recently standardised methodology [31,40]. The heterogeneity between the identified studies precluded meta-analysis and calculation of combined estimates of prevalence and/or incidence rates of neuropathic pain. Nevertheless, we are able to draw comparisons between different studies and conditions and to consider these in the context of the relative quality, strengths and weaknesses, and study designs and reporting.
4.1. Main findings
Overall, the prevalence of neuropathic pain as a feature of chronic pain (or pain of predominantly neuropathic origin) was examined by 8 studies and ranged from 1% to 17.9% (Table 2). In studies that used questionnaire-based case ascertainment tools such as DN4 and S-LANSS, the range of prevalence estimates was wide (3.3–17.9%). However, including only studies that administered case ascertainment tools precisely as they were designed and validated, the range of prevalence rates was much narrower (6.9–10%) [4,8,35,42].
Gustorff et al.  utilised the LANSS (Part A) along with additional pain questions and some specific exclusion criteria, limiting their analysis to people with a known underlying neuropathic pain diagnosis (people with exclusive joint pain or reported rheumatism, migraine, fibromyalgia, visceral pain, and/or satisfying response to over-the-counter analgesic medication were excluded from further interview, as well as people suffering from an undifferentiated cause as the main cause of their pain); this explains their estimate of the lowest prevalence of neuropathic pain at 3.3%.
In contrast, Toth et al.  included all respondents who reported “daily or near daily pain” from telephone surveys of the general population in a Canadian province (finding a prevalence of 38%). Respondents then completed the DN4, with a resultant highest prevalence estimate of neuropathic pain at 17.9%. This is likely an overestimate due to the screening question used to identify the presence of pain or selection bias in the underlying population. Other studies, utilising data obtained from computerised medical records, found a much lower overall prevalence of around 1% [13,14]. This lower rate likely reflects differences in case ascertainment, and in particular that electronic records can only identify patients who have presented to primary care and had a diagnosis recorded by the attending physician.
When we examined studies of specific neuropathic pain conditions, most of them reported incidence rates per person–years, which is a useful parameter for comparing rates, although less clinically useful as a direct indicator of the number of cases likely to arise in a specific population during a specific period. Koopman et al.  consistently reported lower incidence rates of PHN, TN, and GN (Table 3) than the other studies, and this is mostly likely due to differences in case ascertainment. This can be illustrated by comparing this study with that reported by Dieleman et al. . Both of these studies were conducted in consecutive years and both used the IPCI database in the Netherlands, yet they found very different incidence rates. For example, looking at TN, Koopman et al. reported an incidence rate of 12.6/100,000 person–years, whereas Dieleman et al. reported a rate of 28.9/100,000 person–years. Koopman et al. accessed additional information when identifying the cases, and the authors were able to identify and then to exclude misclassified cases (false-positive cases) from the analysis. As a result, the authors suggest their estimate of the incidence of TN was less likely to be exaggerated due to misclassification.
Because of the heterogeneity of included studies, it is not possible to calculate a valid weighted prevalence or incidence for any neuropathic pain condition. Therefore, based on the quality, methods, and other characteristics of the studies we reviewed, and acknowledging this heterogeneity, a best-estimate range of comparable incidence rates for PHN is potentially narrower (28.2–42.0/100,000 person–years) [10,18,19,21], and similar for TN (26.8–28.9/100,000 person–years) [10,18,19], GN (0.4–0.8/100,000 person–years) [10,23], and PLP (0.8–1.5/100,000 person–years) [18,19]. The incidence range of PDPN remains wide (15.3–72.3/100,000 person–years) [10,18,19] despite comparable incidence rates for PHN and TN. Irrespective of these differences, it seems clear that GN and PLP occur much less frequently than PDPN or PHN.
4.2. Other findings
It was surprising that some articles did not provide a working definition for neuropathic pain as a starting point. The articles that did include a definition cited the IASP (Merskey and Bogduk) definition : chronic pain that is “initiated or caused by a primary lesion or dysfunction in the nervous system.” This definition has since been superseded as pain “arising as a direct consequence of a lesion or disease affecting the somatosensory system” [20,25].
Although case ascertainment tools were used by a number of studies, these were neither designed to nor did they claim to make a diagnosis of definite neuropathic pain; rather they were validated in identifying pain with neuropathic characteristics . Some studies were clear in their acknowledgment of this in describing the cases [4,35,38,42], but other studies did not distinguish between neuropathic features and neuropathic pain [8,15].
In studies based on reviews of medical records, the identification of cases relies on accurate diagnoses of neuropathic pain (conditions) by the physician, who then must enter details accurately, using an accurate coding system and an accurate method of retrieving these codes. Rates calculated from medical records also may be affected by severity of pain or disease, as patients with more severe cases are more likely to seek medical attention . Factors that may affect incidence and prevalence rate estimates (in general) are summarised in Table 5.
Although there are published guidelines on the assessment of neuropathic pain , the lack of an agreed-upon specific diagnostic tool has resulted in a proposed grading system of definite, probable, and possible neuropathic pain . This is based on a combination of plausible distribution and history of pain, other positive confirmatory tests, and clinical investigation(s) by a pain specialist or neurologist . For large epidemiological studies, the grading of probable or definite neuropathic pain is not practicable to achieve, and at best the case definitions included in this review may approximate possible neuropathic pain. Existing Special Interest Group on Neuropathic Pain (NeuPSIG) guidelines on neuropathic pain assessment advise that existing screening tools require further validation before they can be completely useful for epidemiological studies in general population samples . Further research should look to explore how the findings of these questionnaire-based screening instruments/case ascertainment tools relate to specialist clinical examinations in a general population sample.
Neuropathic pain is associated with a number of clinical conditions and diseases, but only a small number of them were examined in the articles identified for inclusion in this review. Particular aetiologies such as PHN were more commonly studied, whereas others such as spinal cord injury  were not studied in a general population context. Published studies of the epidemiology of these less common subgroups tended to be conducted in small samples, in specialist settings, or in secondary care, providing an indication of their likelihood in these subgroups, but not of the overall population burden or risk.
4.3. Strengths and weaknesses
The strengths of this review are in its strict systematic approach and consistency in application of the criteria throughout. A thorough search strategy was developed. The studies included were all of medium to high quality.
Although a number of the screening tools that identify pain with neuropathic characteristics have been translated and validated for use in other languages , a limitation in this review was the exclusion of non-English-language articles, as translation was beyond the available resources.
Unlike other systematic reviews [28,31], there is no recognised quality checklist pertaining to observational studies within the general population. Hence a quality checklist was developed from a well-used standard tool specifically for epidemiological studies . Although it is not a standard instrument for systematic reviews, it does provide a reasonable indicator of quality and validity, allowing rejection of articles or studies of low quality and comparison between studies.
4.4. Implications for research and clinical practice
As the use of screening tools within the published literature increases, it may be possible to restrict a future review to a specified case ascertainment tool, eg, DN4 or S-LANSS. The studies included would therefore have greater consistency, would have similar features in common, and would be more comparable when analysing the results.
Now that there is an agreed-upon definition of neuropathic pain, it is important to acknowledge its limitations and the application of this within large epidemiological studies. There is also a good argument that neuropathic pain should be considered as a Medical Subject Headings (MeSH) term in databases.
A comprehensive literature search and systematic review was performed on the epidemiology of neuropathic pain in the general population. A range of incidence and prevalence rates has been identified. The lack of consensus on a definition, diagnostic criteria, and appropriate and consistent use of the screening tools are reflected by the varying methods and results of the studies included in this review. Future epidemiological studies into neuropathic pain within the community should take note of these factors, using a standardised approach for identifying pain with possible neuropathic features.
Conflict of interest
B.S. and N.T. have previously received an unrestricted educational grant from Pfizer UK Ltd. for research on neuropathic pain. B.S. has received occasional lecture and consultancy fees, on behalf of his institution, from Pfizer and Napp. The authors otherwise assert no personal pecuniary or other conflict of interest in the writing of this article. No writing assistance or external funding was obtained for this study.
 Bennett MI, Smith BH, Torrance N, Potter J. The S-LANSS score for identifying pain of predominantly neuropathic origin: Validation for use in clinical and postal research. J Pain 2005;6:149-158.
 Bennett MI, Smith BH, Torrance N, Lee AJ. Can pain can be more or less neuropathic? Comparison of symptom assessment tools with ratings of certainty by clinicians. PAIN® 2006;122:289-294.
 Bennett MI, Attal N, Backonja MM, Baron R, Bouhassira D, Freynhagen R, Scholz J, Tölle TR, Wittchen HU, Jensen TS. Using screening tools to identify neuropathic pain
. PAIN® 2007;127:199-203.
 Bouhassira D, Lanteri-Minet M, Attal N, Laurent B, Touboul C. Prevalence
of chronic pain with neuropathic characteristics in the general population. PAIN® 2008;136:380-387.
 Bouhassira D, Attal N, Alchaar H, Boureau F, Brochet B, Bruxelle J, Cunin G, Fermanian J, Ginies P, Grun-Overdyking A, Jafari Schluep H, Lantéri-Minet M, Laurent B, Mick G, Serrie A, Valade D, Vicaut E. Comparison of pain syndromes associated with nervous or somatic lesions and development of a new neuropathic pain
diagnostic questionnaire (DN4). PAIN® 2005;114:29-36.
 Choo PW, Galil K, Donahue JG, Walker AM, Spiegelman D, Platt R. Risk factors for postherpetic neuralgia. Arch Intern Med 1997;157:1217-1224.
 Davies M, Brophy S, Williams R, Taylor A. The prevalence
, severity, and impact of painful diabetic peripheral neuropathy in type 2 diabetes. Diabetes Care 2006;29:1518-1522.
 De Moraes Vieira EB, Garcia JB, Da Silva AA, Mualem Araujo RL, Jansen RC. Prevalence
, characteristics, and factors associated with chronic pain with and without neuropathic characteristics in Sao Luis, Brazil. J Pain Symptom Manage 2012;44:239-251.
 Di Luzio Paparatti U, Arpinelli F, Visona G. Herpes zoster and its complications in Italy: An observational survey. J Infect 1999;38:116-119.
 Dieleman JP, Kerklaan J, Huygen FJ, Bouma PA, Sturkenboom MC. Incidence
rates and treatment of neuropathic pain
conditions in the general population. PAIN® 2008;137:681-688.
 Dworkin RH. An overview of neuropathic pain
: Syndromes, symptoms, signs, and several mechanisms. Clin J Pain 2002;18:343-349.
 Dworkin RH, Panarites CJ, Armstrong EP, Malone DC, Pham SV. Is treatment of postherpetic neuralgia in the community consistent with evidence-based recommendations? PAIN® 2012;153:869-875.
 Gajria C, Murray J, Birger R, Banarsee R, Bennett DL, Tan K, Field M, Rice AS, Majeed A. Identification of patients with neuropathic pain
using electronic primary care records. Inform Prim Care 2011;19:83-90.
 Gore M, Dukes E, Rowbotham DJ, Tai K, Leslie D. Clinical characteristics and pain management among patients with painful peripheral neuropathic disorders in general practice. Eur J Pain 2007;11:652-664.
 Gustorff B, Dorner T, Likar R, Grisold W, Lawrence K, Schwarz F, Rieder A. Prevalence
of self-reported neuropathic pain
and impact on quality of life: A prospective representative survey. Acta Anaesthesiol Scand 2008;52:132-136.
 Haanpää M, Attal N, Backonja M, Baron R, Bennett M, Bouhassira D, Cruccu G, Hansson P, Haythornthwaite JA, Iannetti GD, Jensen TS, Kauppila T, Nurmikko TJ, Rice AS, Rowbotham M, Serra J, Sommer C, Smith BH, Treede RD. NeuPSIG guidelines on neuropathic pain
assessment. PAIN® 2011;152:14-27.
 Haanpää ML, Backonja MM, Bennett MI, Bouhassira D, Cruccu G, Hansson PT, Jensen TS, Kauppila T, Rice AS, Smith BH, Treede RD, Baron R. Assessment of neuropathic pain
in primary care. Am J Med 2009;122:S13-21.
 Hall GC, Carroll D, McQuay HJ. Primary care incidence
and treatment of four neuropathic pain
conditions: A descriptive study, 2002–2005. BMC Fam Pract 2008;9:26.
 Hall GC, Carroll D, Parry D, McQuay HJ. Epidemiology
and treatment of neuropathic pain
: The UK primary care perspective. PAIN® 2006;122:156-162.
 Jensen TS, Gebhart GF. New pain terminology: A work in progress. PAIN® 2008;140:399-400.
 Jih JS, Chen YJ, Lin MW, Chen YC, Chen TJ, Huang YL, Chen CC. Epidemiological features and costs of herpes zoster in Taiwan: A national study 2000 to 2006. Acta Derm Venereol 2009;89:612-616.
 Katusic S, Beard CM, Bergstralh E, Kurland LT. Incidence
and clinical features of trigeminal neuralgia, Rochester, Minnesota, 1945–1984. Ann Neurol 1990;27:89-95.
 Katusic S, Williams DB, Beard CM, Bergstralh E, Kurland LT. Incidence
and clinical features of glossopharyngeal neuralgia, Rochester, Minnesota, 1945–1984. Neuroepidemiology 1991;10:266-275.
 Koopman JS, Dieleman JP, Huygen FJ, de Mos M, Martin CG, Sturkenboom MC. Incidence
of facial pain in the general population. PAIN® 2009;147:122-127.
 Loeser JD, Treede R. The Kyoto protocol of IASP Basic Pain Terminology. PAIN® 2008;137:473-477.
 MacDonald BK, Cockerell OC, Sander JW, Shorvon SD. The incidence
and lifetime prevalence
of neurological disorders in a prospective community-based study in the UK. Brain 2000;123:665-676.
 Merskey H, Bogduk N. Descriptions of chronic pain syndromes and definitions of pain terms. Classification of chronic pain. Seattle: IASP Press; 1994.
 Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA, Group. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. J Clin Epidemiol 2009;62:1006-1012.
 Opstelten W, Mauritz JW, de Wit NJ, van Wijck AJ, Stalman WA, van Essen GA. Herpes zoster and postherpetic neuralgia: Incidence
and risk indicators using a general practice research database. Fam Pract 2002;19:471-475.
 Petticrew M, Roberts H. How to appraise the studies: An introduction to assessing study quality. Systematic reviews in the social sciences: A practical guide. Oxford: Blackwell Publishing; 2006. p. 125–163.
 Pierik JG, Gumbs PD, Fortanier SA, Van Steenwijk PC, Postma MJ. Epidemiological characteristics and societal burden of varicella zoster virus in the Netherlands. BMC Infect Dis 2012;12:110.
 Siddall PJ, McClelland JM, Rutkowski SB, Cousins MJ. A longitudinal study of the prevalence
and characteristics of pain in the first 5 years following spinal cord injury. PAIN® 2003;103:249-257.
 Smith BH, Torrance N, Ferguson JA, Bennett MI, Serpell MG, Dunn KM. Towards a definition of refractory neuropathic pain
for epidemiological research. An international Delphi survey of experts. BMC Neurol 2012;12:29.
 Torrance N, Smith BH, Bennett MI, Lee AJ. The epidemiology
of chronic pain of predominantly neuropathic origin. Results from a general population survey. J Pain 2006;7:282-289.
 Torrance N, Smith BH, Watson MC, Bennett MI. Medication and treatment use in primary care patients with chronic pain of predominantly neuropathic origin. Fam Pract 2007;24:481-485.
 Torrance N, Ferguson JA, Afolabi E, Bennett MI, Serpell MG, Dunn KM, Smith BH. Neuropathic pain
in the community: More under-treated than refractory? PAIN® 2013;154:690-699.
 Toth C, Lander J, Wiebe S. The prevalence
and impact of chronic pain with neuropathic pain
symptoms in the general population. Pain Med 2009;10:918-929.
 Treede RD, Jensen TS, Campbell JN, Cruccu G, Dostrovsky JO, Griffin JW, Hansson P, Hughes R, Nurmikko T, Serra J. Neuropathic pain
: Redefinition and a grading system for clinical and research purposes. Neurology 2008;70:1630-1635.
 Von Elm E, Altman DG, Egger M, Stuart J, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. Strengthening the reporting of observational studies in epidemiology
(STROBE) statement: Guidelines for reporting observational studies. BMJ 2007;335:7624.
 Yawn BP, Wollan PC, Weingarten TN, Watson JC, Hooten WM, Melton LJ 3rd. The prevalence
of neuropathic pain
: Clinical evaluation compared with screening tools in a community population. Pain Med 2009;10:586-593.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.pain.2013.11.013.
Keywords:© 2014 International Association for the Study of Pain
Epidemiology; Incidence; Neuropathic pain; Prevalence; Systematic review