A systematic literature review of the epidemiology and burden of herpes zoster in selected locales in Asia Pacific

ABSTRACT

Herpes zoster (HZ) is a painful rash which typically affects older adults. This is of concern in Asia-Pacific given its aging population. As HZ epidemiology and burden are evolving, this systematic literature review aimed to update the current understanding of HZ burden and associated costs for selected Asia-Pacific locales. MEDLINE and Embase were searched for English articles of HZ studies conducted in Australia, China, Hong Kong, Japan, Korea, New Zealand, Singapore, and Taiwan. Eligible outcomes included HZ incidence and prevalence, occurrence of HZ-related complications, healthcare resource utilization, costs, and HZ-associated quality of life outcomes. This paper focused on HZ data in the general adult population (N = 90 articles). Substantial HZ-related disease and economic burden were observed in these locales, consistent with global trends. These findings reinforce the increasing burden of HZ and need for preventive strategies, which may include raising awareness and encouraging timely vaccination.

Plain Language Summary

Herpes zoster, also known as shingles, is a painful rash that usually resolves after a few weeks, although some people experience serious or long-lasting complications. Shingles is common, affecting around one in every three individuals in their lifetime, and older persons are more likely to have shingles. Given the aging population in the Asia-Pacific region, shingles represents an increasingly important health issue as the proportion of older people increases. Vaccination can help prevent shingles and avoid its complications. New data on the trends and burden of shingles in this region are regularly generated. Therefore, in this study, we looked at studies from selected countries published over the past twenty years to summarize the latest available information on: how many people experience shingles in selected Asia-Pacific areas, how these individuals and societies are affected, and the related costs. Consistent with previous research, this study observed an increasing trend in the number of persons with shingles and costs of managing it, especially in older adults. In populations that are aging, there is a need for ways to reduce the risk of shingles and to lessen its burden on the healthcare system and society. Our findings can help to inform current development of strategies to reduce the risk of shingles, including education (on the burden and risk of shingles) and encouraging uptake of preventive measures.

KEYWORDS:

• Herpes zoster
• Asia-Pacific
• epidemiology
• complications
• healthcare burden
• shingles
• systematic literature review

Introduction

Herpes zoster (HZ), also known as shingles, occurs due to the reactivation of the varicella zoster virus (VZV).^1,2 HZ can cause complications which are challenging to treat, the most common and debilitating being postherpetic neuralgia (PHN).^1,3 HZ and its complications negatively impact the quality of life (QoL) of patients, posing significant individual burden as well as economic burden on society and a strain on healthcare resources.^3,4

HZ incidence and the risk of complications have been found to increase substantially in older adults after 50 years of age (YOA).^1,2 The risk of HZ increases with a decline in immune system function that is often due to aging (immunosenescence) or immunocompromised conditions.^1,2 This presents a challenging health issue in the Asia-Pacific region which has a rapidly aging population. By 2050, the number of persons aged ≥60 years in the Asia-Pacific region is expected to reach 1.3 billion, representing approximately 25% of the population.⁵ Data from the Asia-Pacific region have shown VZV seroprevalence rates of up to > 90% in adults^4,6; as VZV seropositivity indicates VZV latency, this suggests that potentially almost every adult is at risk of VZV reactivation, and hence HZ, in their lifetime.

The substantial health and economic challenges posed by HZ and its complications have driven the development of two types of HZ vaccines: the live attenuated zoster vaccine (ZVL) and the adjuvanted recombinant zoster vaccine (RZV). The first ZVL was approved by the United States Food and Drug Administration (FDA) in 2006 but only became widely commercially available after 2011 including in Asia-Pacific,^3,6–10 while RZV was FDA-approved in 2017 and first launched in the Asia-Pacific region in 2020.^11,12

A comprehensive systematic literature review (SLR) in the Asia-Pacific region has observed an average HZ incidence rate of 3–10/1,000 person-years (PY), however only data up to 2014 were utilized and they do not provide insight from studies over the past decade.⁶ Other more recent SLRs and meta-analyses have focused on specific regions such as Southeast Asia only,¹³ or on specific populations (e.g., adults ≥ 50 YOA) with Asia-Pacific data described as a part of worldwide results.^14,15

Given the above considerations, the present SLR aimed to provide an updated summary of evidence on HZ epidemiological and disease burden in the Asia-Pacific region, based on data up to 2022, with an overarching goal to address current gaps in the literature and guide future studies to inform HZ vaccination strategies. HZ vaccine distribution within the Asia-Pacific region has primarily occurred in more developed areas,¹⁰ hence this SLR focused on Asia-Pacific locales with advanced economies and relatively higher per capita gross domestic product (GDP).¹⁶ Moreover, HZ epidemiology data are limited in low- and middle-income locales.^1,10 This paper specifically reports the findings related to the general adult population from articles identified through the SLR.

Methods

Information sources and search strategy

This SLR was performed with reference to the Preferred Reporting Items for Systematic Literature Reviews and Meta-Analyses (PRISMA) guidelines to obtain relevant information using reproducible and transparent methodology.¹⁷ Articles published in the English language from January 1, 2000 to October 1, 2020 were identified on MEDLINE and Embase databases. A subsequent database search using the same search strategy (targeted literature review; TLR) was conducted to identify new studies published from October 1, 2020 to April 30, 2022 on the MEDLINE database only. Reference lists of identified articles were hand-searched to supplement the database searches.

The following search algorithm was applied using all MeSH terms in all fields: [(herpes zoster OR shingles OR postherpetic neuralgia) AND (China OR Japan OR Asia OR Asia Pacific OR Hong Kong OR Taiwan OR Korea OR Singapore OR Australia OR New Zealand) AND (incidence OR prevalence OR seroprevalence OR epidemiology OR complication OR morbidity OR cost OR expenditure OR economic OR burden OR recurrence OR mortality OR death OR healthcare OR inpatient OR outpatient OR hospital OR medical OR physician OR quality of life OR QALY OR knowledge OR attitudes)].

Selection process and screening

The inclusion criteria were: 1) studies published in the English language, 2) studies providing HZ data for adults ≥18 years, 3) studies conducted in Australia, China, Hong Kong, Japan, Korea, New Zealand, Singapore, and Taiwan, and 4) observational studies. Case reports, clinical trials, meta-analyses, reviews, and letters to the editor were excluded. Articles that did not report on the outcomes of interest were also excluded. The outcomes of interest in this SLR were: HZ epidemiology (incidence, prevalence, hospitalization, mortality, complications), HZ-related healthcare resource utilization (HCRU) or costs, HZ-related QoL factors, and HZ-related knowledge, attitudes, and practices (KAP).

Two independent reviewers screened the titles and abstracts of all identified articles for inclusion; an article was included for full-text review if it was found to meet the inclusion criteria by at least one reviewer. One reviewer extracted the data from each full text into a structured data abstraction form. All extracted data were then reviewed and verified by the second reviewer.

Data extraction and collection

The following data related to HZ were extracted: incidence, prevalence, incidence of hospitalizations, mortality, PHN or other complications (e.g., dermatological, ocular), medical costs, HCRU, QoL, patient‐reported burden (e.g., work impact), and KAP (from patients, healthcare workers, or the general population) related to HZ or HZ vaccination. Data extraction parameters also included publication details, locale, study characteristics (design, period), population characteristics (setting, age, sex, race, underlying conditions), and case definitions (HZ and PHN).

Reporting quality appraisal

The completeness and adequacy of the reporting for each study was assessed using the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) checklist for observational studies.¹⁸

Cost conversions

To facilitate comparisons between different locales and to account for inflation, all HZ-related cost estimates were adjusted to 2022 values based on the worldwide inflation rate (per locale) from the World Bank Database,¹⁹ then converted to United States Dollars (USD), if the cost was reported in another currency, based on 2022 currency conversion rates obtained from the United Nations Conference on Trade and Development (UNCTAD) database.²⁰

Results

The initial database searches returned 4,501 articles. Following deduplication and initial title screening for population, locale and study design, a total of 1,669 articles were screened at abstract review, and of these, 387 articles were reviewed in full, and 220 articles were eligible for inclusion in the study. In the subsequent targeted literature review, 73 articles were reviewed in full, and 46 articles were eligible for inclusion in the study. Together with articles identified through hand-searching of reference lists, a total of 271 articles were included in the study (Figure 1).

Figure 1. PRISMA flowchart detailing the number of records identified from searches and included in this manuscript. The number of records identified through database searches are combined for the systematic literature review and targeted literature review. HZ: herpes zoster; HCRU: healthcare resource utilization; PRISMA: preferred reporting items for systematic reviews and meta-Analyses; SLR: systematic literature review; TLR: targeted literature review.

This manuscript reports HZ epidemiology and disease burden in the general adult population and synthesizes information from 90 of the 271 articles identified in the study. The remaining articles reported information pertaining to KAP findings and comorbid populations, which would be reported elsewhere.

Quality appraisal

Overall, the majority of studies included in this SLR adequately addressed the STROBE checklist items. Most HZ burden studies included in this SLR harnessed expansive population-based databases, with a notable concentration in Australia, Japan, Korea and Taiwan. The deliberate selection of large-scale population samples in these studies was instrumental in ensuring statistical power suitably robust to discern significant associations. Additionally, the utilization of nationwide databases in these investigations and enrollment of demographically representative study cohorts extend the external validity of findings from the current SLR.

Population-based disease surveillance and cohort studies crucially contributed toward the reliability of the SLR findings. Nevertheless, the variances in study duration, case definition and case attainment introduced some heterogeneity. Notably, analyses based on databases are inevitably influenced by the nature of the data source (e.g., electronic medical records [EMR], insurance claims, administration records), as well as the breadth of the database which could pertain to the entire population or solely hospital-centric data.

Moreover, studies based on community-based surveys bear inherent limitations, including the potential for recall bias and the absence of definitive diagnostic confirmation. HZ burden studies in clinical and hospital settings, compared with their larger-scale counterparts, encompass smaller sample sizes, potentially limited generalizability, and sometimes lack comprehensive definitions of methodological components (e.g., HZ definition).

Despite the comprehensive scope of this appraisal, there are some items on the STROBE checklist that have not been uniformly addressed across many studies in this SLR. These unaddressed checklist items pertain to the transparency in describing methods for tackling potential bias and accentuating the generalizability of results.

HZ epidemiology in the general population

The incidence of HZ was estimated in studies that adopted various research approaches including population-based assessments (e.g., cohort studies, ecological studies). These results are summarized in Table 1A, and Figures 2 and 3.

Figure 2. Trends across age groups for (a) HZ incidence overall and (b) HZ incidence by sex across selected locales in Asia-Pacific. Open circles represent data for the specified age group and above (i.e., ≥70, ≥80 years). Only data points with values > 0.1 have been included. (a) For studies that reported HZ incidence for multiple periods, data for each period are represented by individual lines, with the period indicated after the study name. (b) Data are presented collectively for all locales included in the SLR and differentiated only by male and female to focus on overall potential sex-specific trends^.aFirst data point was for age group 18–30 years; ^bFinal data point for 70–79 years was from year 2007. HZ: herpes zoster; PY: person-years. Articles included: Australia,^21–23 China,²⁴ Japan,^25–29 Korea,^30–32 and Taiwan.³³

Figure 3. Trends over time for (a) HZ incidence overall, (b) HZ incidence by sex, and (c) all HZ-related hospitalization rates across selected locales in Asia-Pacific. Only data for the overall population (all ages) have been included. Filled circles and solid lines represent data specific to the respective year and trend, respectively; open circles and dashed lines represent an average value over the period specified. (c) Hospitalization rates included all patients with HZ-related diagnosis (principal or non-principal), unless otherwise stated. ^aHospitalization for principal HZ diagnosis only; ^bStudy investigated HZ hospitalization rates in periods based on the availability and funding status of varicella vaccination in Australia; ^cStudy did not specify if hospitalization data included patients with principal and/or non-principal HZ diagnosis; ^dHospitalization for nonindigenous Australians up to age 70 years only. HZ: herpes zoster; PY: person-years. Articles included: Australia,^34–38 China,²⁴ Japan,^{25,28,29,39,40} Korea,^{30,31,41–44} New Zealand,⁴⁵ and Taiwan.^33,46–48

Table 1A. HZ incidence and recurrence rates across selected locales in Asia-Pacific (study-specific findings).

Locale	Study design	Incidence in adults aged ≥50 years per 1,000 PY [age; period]	Incidence in overall population per 1,000 PY [all ages unless specified; period]	HZ recurrence per 1,000 PY or otherwise specified [age; period]
Australia	Prospective cohort	Liu 2015^Citation21 12.1 [>80; 2006–09] Qian 2021^Citation50 9.4^a [≥45; 2004–15]	NR	Qian 2021^Citation50 11.1 [≥45; 2004–15]
	Retrospective cohort	Lin 2021^Citation23 7.5 [70–79; 2017] 5.3 [70–79; 2018] Lin 2022^b,Citation22 7.7 [60–69; 2013–16] 8.3 [60–69; 2016–18] 9.5 [70–79; 2013–16] 7.5 [70–79; 2016–18] 9.0 [80–89; 2013–16] 9.3 [80–89; 2016–18]	NR	NR
China	Cross-sectional survey	Zhu 2015^Citation59 4.3 [≥50; 2011–13] 4.1 [≥50; 2011] 3.4 [≥50; 2012] 5.8 [≥50; 2013]	NR	NR
	Retrospective cohort	Sun 2021^Citation54 6.6 [≥50; 2015–17] 6.1 [≥50; 2015] 6.5 [≥50; 2016] 7.4 [≥50; 2017] 5.1 [50–54; 2015–17] 7.0 [55–59; 2015–17] 7.7 [60–64; 2015–17] 8.4 [65–69; 2015–17] 8.6 [70–74; 2015–17] 7.8 [75–79; 2015–17] 5.7 [80–84; 2015–17] 2.1 [≥85; 2015–17] Li 2022^Citation24 0.5 [50–59; 2013] 0.6 [50–59; 2014] 1.1 [50–59; 2015] 1.5 [50–59; 2016] 1.7 [50–59; 2017] 0.7 [60–69; 2013] 0.9 [60–69; 2014] 1.7 [60–69; 2015] 2.1 [60–69; 2016] 2.5 [60–69; 2017] 1.1 [≥70; 2013] 1.4 [≥70; 2014] 2.4 [≥70; 2015] 3.2 [≥70; 2016] 3.2 [≥70; 2017]	Li 2022^Citation24 0.3 [2013] 0.4 [2014] 0.7 [2015] 1.0 [2016] 1.2 [2017]	Li 2016^Citation53 2.8%^c [≥50; 2010–12] Sun 2021^Citation54 16.6 [≥50; 2015–17] 13.8 [50–54; 2015–17] 9.4 [55–59; 2015–17] 13.1 [60–64; 2015–17] 20.9 [65–69; 2015–17] 24.6 [70–74; 2015–17] 28.1 [75–79; 2015–17] 36.7 [80–84; 2015–17] 20.9 [≥85; 2015–17]
	Retrospective survey	NR	NR	Lu 2018^Citation52 11.1%^d [>0; 2012–13]
Japan	Ecological study	Toyama 2018^Citation39 6.2 [>60; 1997] 9.5 [>60; 2017]	Toyama 2018^Citation39 3.6 [1997] 6.0 [2017]	NR
	Prospective cohort	Toyama 2009^Citation29 5.2 [50–59; 1997–2006] 7.0 [60–69; 1997–2006] 7.8 [70–79; 1997–2006] 6.9 [80–89; 1997–2006] 5.4 [≥90; 1997–2006] Takao 2015^Citation27 10.9 [≥50; 2008–12] 9.2 [50–59; 2008–12] 9.6 [60–69; 2008–12] 12.9 [70–79; 2008–12] 12.6 [≥80; 2008–12] Nakamura 2016^Citation58 11.0 [≥50; 2008–11] Sato 2017^Citation26 10.2 [≥60; 2013–15] 9.2 [60–69; 2013–15] 10.4 [70–79; 2013–15] 12.1 [≥80; 2013–15] Yamada 2019^Citation81 5.6 [≥50; 2009–12] Kawahira 2022^Citation120 9.3/12.2/8.2^e [≥50; 2008–09]	Toyama 2009^Citation29 4.2 [1997–2006]	Nakamura 2016^Citation58 10.1 [≥50; 2008–11]
	Retrospective cohort	Shiraki 2017^Citation25 8.7 [70–80; 2009–15] Shiraki 2021^Citation28 5.2 [50–59; 1997] 5.2 [50–59; 2014] 6.7 [50–59; 2020] 6.2 [60–69; 1997] 7.8 [60–69; 2014] 8.8 [60–69; 2020] 6.5 [70–79; 1997] 9.5 [70–79; 2014] 10.5 [70–79; 2020] 5.7 [80–89; 1997] 8.6 [80–89; 2014] 10.0 [80–89; 2020] 4.7 [≥90; 1997] 6.4 [≥90; 2014] 7.6 [≥90; 2020]	Shiraki 2017^Citation25 4.8 [2009–15] Imafuku 2019^Citation70 4.9 [≥18 YOA; 2005–14] Kobayashi 2019^Citation121 1.5 [2005–16] Shiraki 2021^Citation28 3.6 [1997] 5.2 [2014] 6.5 [2020]	Shiraki 2017^Citation25 0.3 [>0; 2009–15]
Korea	Cross-sectional survey	NR	NR	Ha 2017^Citation122 1.2%^c [>0; 2005–15]
	Prospective cohort	Kwon 2016^Citation31 16.7 [50–60; 2003–13] 17.8 [60–70; 2003–13] 14.2 [>70; 2003–13]	Kwon 2016^Citation31 9.9 [≥18 YOA; 2003–13]
	Retrospective cohort	Kim 2014^Citation30 17.4 [50–59; 2011] 22.4 [60–69; 2011] 21.8 [70–79; 2011] 16.5 [≥80; 2011] Bae 2020^Citation32 16.7 [≥50; 2002–13] 16.1 [50–59; 2002–13] 18.2 [60–69; 2002–13] 15.4 [≥70; 2002–13] Cho 2020^Citation123 10.7–14.5^a [40–79; 2004–13]	Kim 2014^Citation30 10.4 [2011] Choi 2019^Citation41 4.2 [2006] 9.2 [2015] Choi 2019^Citation42 2.7 [2003] 3.0 [2004] 3.5 [2005] 4.0 [2006] 5.5 [2007] 6.8 [2008] 7.3 [2009] 7.7 [2010] 8.3 [2011] 9.0 [2012] 9.5 [2013] 9.7 [2014] 9.8 [2015] Kim 2019^Citation43 5.1 [>20 YOA; 2002–13]	Kim 2019^Citation43 12.0 [>20; 2002–13] 5.3%^c [>20; 2002–13] 4.5%^c [20–50; 2002–13] 5.8%^c [>50; 2002–13]
Taiwan	Ecological study	NR	Wu 2013^Citation48 4.0 [2000] 6.2 [2009]	NR
	Retrospective cohort	Chao 2012^Citation33 8.1 [50–59; 2000] 12.1 [50–59; 2008] 10.5 [60–69; 2000] 14.9 [60–69; 2008] 12.5 [70–79; 2000] 17.2 [70–79; 2007] Lu 2018^Citation46 9.2 [50–64; 2004] 9.5 [50–64; 2005] 9.6 [50–64; 2006] 9.7 [50–64; 2007] 10.0 [50–64; 2008] 12.3 [65–74; 2004] 12.1 [65–74; 2005] 12.1 [65–74; 2006] 12.5 [65–74; 2007] 12.8 [65–74; 2008]	Jih 2009^Citation47 4.9 [2000–06] Chao 2012^Citation33 4.5 [2000] 4.6 [2001] 5.4 [2002] 5.3 [2003] 6.0 [2004] 6.2 [2005] 6.5 [2006] 6.6 [2007] 6.9 [2008] Lu 2018^Citation46 5.0 [2004] 5.3 [2005] 5.3 [2006] 5.4 [2007] 5.7 [2008]	NR
		12.6 [75–84; 2004] 12.6 [75–84; 2005] 13.0 [75–84; 2006] 13.1 [75–84; 2007]
		13.5 [75–84; 2008] 11.2 [≥85; 2004] 11.4 [≥85; 2005] 11.0 [≥85; 2006] 11.8 [≥85; 2007] 12.4 [≥85; 2008]

^aData were only available for the age group presented which mostly consists of and may reasonably represent the data of adults aged ≥50 years; ^bData reported for before (Nov 2013–Oct 2016) and after (Nov 2016–Oct 2018) national HZ immunization program; ^cProportion of incident HZ cases with recurrence; ^dProportion of interviewed HZ patients with recurrent HZ; ^eIncidence according to BMI status under/normal/overweight. BMI: body mass index; HZ: herpes zoster; PY: person-years; NR: not reported; YOA: years of age.

HZ incidence

The estimates of HZ incidence rate in the overall general population (all ages; 0.3–10.4/1,000 PY) and in adults ≥ 50 YOA (3.4–16.7/1,000 PY) varied by locale, study design, age group, and over time (Table 1A). A one-third lifetime risk of HZ was reported in Japan²⁵ and Taiwan.⁴⁹ Population-based assessments consistently found that the incidence of HZ generally increased with age in Australia,^{21,22,34,50,51} China,^24,52–54 Japan,^25–29,39 Korea,^30,31,41 New Zealand,⁵⁵ and Taiwan,^{33,46,47,49,56} especially after 50 YOA^{25,30,31,33,34,41,46,51,52,55} (Figure 2A). Notably, overall HZ incidence rates were comparatively higher in Korea.^30–32 HZ incidence rates generally declined after 70 YOA, in studies where such data were available (Figure 2A). The incidence of HZ was observed to be higher in females than males in Australia,^21,50 China,^52–54 Japan,^{25–29,57,58} Korea,^30,41 New Zealand,⁵⁵ and Taiwan^48,49 (Figure 2B).

Furthermore, an increasing trend of HZ incidence over time was reported in Australia,³⁴ China,^24,54,59 Japan,^28,29,39 Korea,^41,42 and Taiwan^33,46,48,56 (Figure 3A). Increases in HZ notification and HZ-related consultation over time were also reported in Australia^34,35,60,61 and New Zealand.⁴⁵ More recent surveillance reports in Australia however observed that HZ-related consultation rates were comparable from 2013–2017.^62–66

HZ recurrence

The reported incidence of recurrent HZ per 1,000 PY varied by locale and age group: 11.1 in Australia (≥45 YOA),⁵⁰ 10.1 in Japan (≥50 YOA),⁵⁸ 12.0 in Korea (>20 YOA),⁴³ and 16.6 in China (≥50 YOA)⁵⁴ (Table 1A). Generally, the incidence of HZ recurrence increased with age in Australia,⁵⁰ China,⁵⁴ and Korea.⁴³ The incidence of recurrent HZ was lower than that of initial HZ in Japan,²⁵ but higher than the incidence of initial HZ in Australia,⁵⁰ China,⁵⁴ and Korea⁴³ within the individual study cohorts (follow-up periods varied from 3 to 10 years). There were inconsistent findings on the incidence of recurrent HZ in males versus females, although a greater proportion of studies reported higher recurrence in females.^25,43,50

HZ-related hospitalization

Estimates of HZ-related hospitalization in the general population, reflecting the burden of disease, are summarized in Table 1B. HZ-related hospitalization rates were found to increase over time (Figure 3C) in Australia,^34–36 Japan,⁴⁰ and Korea,^41,44 and increase with age in Australia,^{21,34,35,37,51,67} Hong Kong,⁶⁸ and Korea.⁴¹

Table 1B. HZ-related hospitalization rates across selected locales in Asia-Pacific (study-specific findings).

Locale	Study design	Hospitalization in adults aged ≥50 years per 100,000 or otherwise specified [age; period]	Hospitalization in overall population per 100,000 or otherwise specified [all ages unless specified; period]
Australia	Ecological study	Heywood 2014^Citation37 17.6 [50–59; 1998–99] 17.0 [50–59; 2000–03] 18.3 [50–59; 2004–05] 16.4 [50–59; 2006–10] 46.3 [60–69; 1998–99] 38.7 [60–69; 2000–03] 42.3 [60–69; 2004–05] 42.4 [60–69; 2006–10] 129.3 [70–79; 1998–99] 120.2 [70–79; 2000–03] 116.5 [70–79; 2004–05] 105.9 [70–79; 2006–10] 296.1 [≥80; 1998–99] 271.0 [≥80; 2000–03] 280.4 [≥80; 2004–05] 271.5 [≥80; 2006–10]	Heywood 2014^Citation37 24.8 [1998–99] 23.9 [2000–03] 25.2 [2004–05] 24.5 [2006–10]
	Prospective cohort	Liu 2015^Citation21 0.9/1,000 PY [>80; 2006–11]	NR
	Retrospective cohort	Stein 2009^Citation51 67 [≥50; 1998–2005]	NR
	Retrospective review	Carville 2010^Citation36 10.0 [50–59; 1995–99] 12.1 [50–59; 2000–05] 12.3 [50–59; 2006–07] 25.2 [60–69; 1995–99] 30.3 [60–69; 2000–05] 35.4 [60–69; 2006–07] 63.3 [70–79; 1995–99] 89.9 [70–79; 2000–05] 91.1 [70–79; 2006–07] 148.4 [≥80; 1995–99] 187.8 [≥80; 2000–05] 214.2 [≥80; 2006–07] MacIntyre 2015^Citation34 19 [50–59; 1998–2005] 18 [50–59; 2007–12] 41 [60–69; 1998–2005] 43 [60–69; 2007–12] 123 [70–79; 1998–2005] 112 [70–79; 2007–12] 267.3 [>80; 1998–2005] 289.1 [>80; 2007–12] Sheel 2017^a,Citation38 8.3 [50–59; 2007–11] 18.8 [60–69; 2007–11] 44.8 [70–79; 2007–11] 89.8 [≥80; 2007–11] Sheel 2018^Citation35 17.9 [50–59; 1999–2013] 43.2 [60–69; 1999–2013] 90.2 [70–74; 1999–2013] 228.1 [≥75; 1999–2013]	Carville 2010^Citation36 13.4 [1995–99] 18.2 [2000–05] 20.5 [2006–07] MacIntyre 2015^Citation34 24.6 [1998–2005] 26.7 [2007–12] Sheel 2017^a,b,Citation38 11.2 [2007–11] Sheel 2018^Citation35 25.9 [1999–2013]
Hong Kong	Retrospective review	Chan 2018^Citation68 117 [≥60; 2004–15]	NR
	Self-controlled case series	NR	Wan 2022^Citation124 7.9 per 1 million BNT162b2 doses administered^c [≥16 YOA; 2021] 7.1 per 1 million CoronaVac doses administered^c [≥18 YOA; 2021]
Japan	Cross-sectional survey	Kuniyoshi 2021^d,Citation40 35.3 [61–79; 2013–18] 63.2 [≥80; 2013–18]	Kuniyoshi 2021^d,62 17.1 [2013] 17.5 [2014] 18.4 [2015] 18.2 [2016] 18.3 [2017] 18.4 [2018]
Korea	Retrospective cohort	Cho 2020^Citation123 0.6–1.2 per 1,000 PY [40–79^e; 2004–13]	Choi 2010^a,Citation44 22 [2003] 25 [2004] 27 [2005] 30 [2006] 32 [2007]
Taiwan	Retrospective cohort	NR	Jih 2009^a,Citation47 16.1/100,000 PY [2000–06]

Table presents data for all HZ-related hospitalizations (principal and non-principal HZ diagnoses), unless otherwise stated. ^aHospitalization for principal HZ diagnosis only; ^bHospitalization for non-indigenous Australians up to age 70 years only; ^cHospitalization with principal diagnosis of HZ within 28 days of vaccination; ^dStudy did not specify if hospitalization data included patients with principal and/or non-principal HZ diagnosis; ^eData were only available for the age group presented which mostly consists of and may reasonably represent the data of adults aged ≥50 years. HZ: herpes zoster; PY: person-years; NR: not reported.

Impact of HZ vaccination on HZ occurrence

Limited studies evaluated the impact of HZ vaccination on HZ occurrence and data were only available from Australia. Following implementation of the Australian National HZ Immunization Program (live-attenuated HZ vaccine for adults 70–79 YOA), HZ incidence in the target age group was observed to decrease from 9.5/1,000 PY (2013–2016) to 7.5/1,000 PY (2016–2018),²² and was lower in vaccinated patients than unvaccinated patients (3.6–3.9/1,000 PY versus 6.3–8.7/1,000 PY).²³

HZ complications and mortality

The complications associated with HZ are summarized in Figure 4. The most common HZ complication reported was PHN, and data on PHN among HZ patients were available for all locales except New Zealand. There was a broad range of reported data on PHN, partly due to the heterogenous definition of PHN used in the studies, which generally differed by diagnostic criteria and duration of pain. For consistency, where studies have used more than one PHN definition (e.g., outcomes reported for patients with different durations of pain), only PHN data based on the conventional definition of dermatomal pain persisting for at least three months from HZ onset are reported in this SLR.^1,93

Figure 4. Proportion of HZ cases with HZ-related complications across selected locales in Asia-Pacific. Black bars represent data of all HZ patients in the study population; blue bars represent data of a study population comprising hospitalized HZ patients only; orange bars represent data of a study population comprising HZO patients only. ^aPatients with pain persisting for at least 3 months from HZ onset; ^bStudy did not specify the duration of pain in patients diagnosed with PHN; ^cPHN was defined as pain persisting for at least 90 days after HZ rash healed; ^dPatients with persistent pain at 6 months from HZ onset; ^eData pooled from patients in Taiwan, Korea, and Thailand; ^fOther complications include motor dysfunction (Tang 2022) or were otherwise not defined. HZ: herpes zoster; HZO: herpes zoster ophthalmicus; PHN: postherpetic neuralgia. Articles included: dermatological/soft tissue infection,^26,49,69 disseminated HZ,^{51,54,68,70,71} ear-related/Ramsay-hunt syndrome,^54,68,71 encephalitis,^51,54,68 lower respiratory tract infection,⁴⁹ meningitis,^51,68,72,73 meningitis encephalitis,⁷¹ neurological,^{26, 51,68–70,72} ocular/ophthalmic,^{26,49,51,54,55,68–72 ,74–76} PHN,^{26,27,35,46,47,52–54,59,68–71,75,77–89} vision loss,^90,91 and other.^{26,51,54,70,72,92}.

The incidence of PHN for all ages ranged from 0.42/1,000 PY in Taiwan⁴⁷ to 2.5/1,000 PY in Korea.⁹⁴ The incidence of PHN was found to increase over time in Australia (0.5/1,000 persons in April 2000–September 2006 to 0.8/1,000 persons in October 2006–March 2013).³⁴ The standardized prevalence of PHN increased annually from 1.6/1,000 persons in 2009 to 2.2/1,000 persons in 2013 in Korea.⁹⁵

Older age was associated with PHN in multiple studies. In older adults ≥ 50 YOA, PHN incidence was 0.5/1,000 PY in China⁵⁴ and 2.1/1,000 PY in Japan.²⁷ In China,⁷⁷ Japan,²⁷ and Korea,⁹⁴ PHN incidence continued to increase in patients > 70 YOA. In Australia, PHN incidence was 1.0/1,000 persons in those aged 50–59 years and 5.3/1,000 persons in those aged ≥80 years.³⁴

While the percentage of all HZ cases (all ages) who developed PHN were 4.0–12.4% in Japan,^70,78 4.6–24.9% in China,^52,79 and 2.4–11.4% in Taiwan,^46,47,80 there was a trend for these percentages to be higher among older patients ≥ 50 YOA, whereby 9.2–19.7% in Japan,^26,27,69,81 7.3–44.8% in China,^53,54,59,82 and 20.7% in Taiwan⁸³ developed PHN.

The percentage of HZ patients who developed PHN was also higher among patients who were hospitalized, whereby 12.8–32.5% of hospitalized HZ patients (all ages) in Australia, China, and Hong Kong developed PHN.^35,68,77 In Hong Kong, the proportion of hospitalized HZ cases with PHN was higher in older adults aged ≥50 years (39.0%) than the overall population (12.8%).^68,71

Other factors associated with PHN were more severe skin lesions,^77,84 severe pain at rash onset or the initial visit,^26,77,85 underlying diabetes,⁸⁴ immunosuppressive therapy,^26,69,96 sleep shortage,⁸¹ and psychosocial factors.⁹⁷ The male sex was associated with PHN development in Japan,²⁶ while in Korea, PHN incidence was observed to be higher in females.⁹⁴

Besides PHN, other HZ-related complications were not well described, and the estimates varied by locales (Figure 4). In Australia, the proportions of various HZ-related complications were similar for all age groups and older patients aged ≥50 years.^51,72 In Hong Kong, similar HZ-related complications were reported in hospitalized HZ cases and those in the Accident and Emergency (A&E) unit, although there was a lower proportion of patients for the latter.⁶⁸ Complications reported in hospitalized HZ patients were generally of the following categories: neurological (33.1–35.1% in Australia, 15.9% in Hong Kong),^51,68,72 ear-related including Ramsay-Hunt syndrome (1.9–3.4% in Hong Kong),^68,71 ocular or ophthalmic (16.0–16.2% in Australia, 4.9–6.7% in Hong Kong, 3.7% in Japan, 11.3% in Taiwan),^{49,51,68,71,72,74} meningitis (0.2–0.8% in Australia, 0.5% in Hong Kong),^51,68,72 encephalitis (1.1% in Australia, 0.2% in Hong Kong),^51,68 meningitis encephalitis (1.0% in Hong Kong),⁷¹ and disseminated HZ (1.1% in Australia, 0.6–1.0% in Hong Kong).^51,68,71 The variability observed in these proportions may be attributed to inconsistent definitions and methods of ascertaining the conditions in patients across the studies.

Reported HZ mortality rates (deaths coded as due to HZ) in Australia were approximately 0.2–0.3/100,000 persons (aged ≥40 years).^35,51 In particular, more than 80% of HZ-related deaths were among patients aged ≥80 years.³⁵ The case fatality rate of hospitalized HZ patients (with zoster as the principal or secondary diagnosis) ranged from 3.8% in Hong Kong to 3.9% in Australia.^68,72 In Japan, 30-day and 60-day survival rates among hospitalized HZ patients were 97.0% and 87.7%, respectively.⁷⁴

Healthcare resource utilization

HZ-related healthcare resource utilization (including the number of outpatient, general practitioner [GP], and emergency department [ED] visits, hospitalization rates and length of stay [LOS]) are detailed in Table 2. No studies were found that reported HCRU pertaining to HZ in New Zealand and Singapore. In general, HCRU of patients with HZ increased with age, and were higher in HZ cases with PHN or other complications.

Table 2. HZ-related healthcare resource utilization in patients with HZ across selected locales in Asia-Pacific (mean per HZ case across period of available data, unless otherwise stated).

Locale	Source	Study period	Age (years)	Patient disease profile	GP visits (n/case)	Outpatient visits (n/case)	Hospitalizations^a (n/case or % cases)	LOS (mean days)	ED visits (n/case or % cases)
Australia	Stein 2009^Citation51	2000–2006^b	≥50	All cases	1.9–3.0	NR	• Principal HZ: 28/10^5 cases • Non-principal HZ: 39/10^5 cases	6.8 (principal HZ)	38.2/10^5 cases
Australia	Liu 2015^Citation21	2006–2009	≥45	All cases	NR	NR	3.3%	NR	NR
Australia	Karki 2016^Citation98	2006–2009 (recruitment) to 2011^b	≥45	All cases^c	1.6 (excess)	2.0 (excess prescriptions filled)	0.06 (excess)	NR	0.11 (excess)
Australia	Qian 2021^Citation50	2004–2015	≥45	All cases	NR	NR	• First HZ episode: 2.5% • Recurrent HZ: 2.7%	NR	NR
China	Zhu 2015^Citation59	2011–2013	≥50	All cases	NR	NR	6.9%^d	NR	NR
China	Li 2016^Citation53	2013–2014	≥50	All cases	NR	NR	4.5%^d	10	NR
China	Lu 2018^Citation52	2012–2013	All ages	All cases	NR	NR	2.3%^d	NR	NR
				Non-PHN	NR	NR	1.9%^d	NR	NR
				PHN	NR	NR	10%^d	NR	NR
China	Yu 2019^Citation99	2016^b	≥40	All cases	NR	9.4	0.7	14.9	0.4
				PHN-related	NR	5.8	0.4	9.2	0.4
China	Sun 2021^Citation54	2015–2017	≥50	w/complications	NR	5.6	2.0	22.5	NR
				w/o complications	NR	3.3	2.3	20.8	NR
Hong Kong	Ming 2019^Citation71	2011–2013	≥50	All cases	NR	NR	NR	7.1^d	NR
Japan	Nakamura 2017^Citation69	2013–2015	≥60	All cases	NR	5.7	1.9%^d	9.4	NR
Japan	Sato 2017^Citation26	2013–2015	≥60	All cases	NR	5.7	3.4%^d,e	NR	NR
Japan	Imafuku 2019^Citation70	2005–2014	≥18	All cases	NR	3.4	2.3%^d	9.1	NR
Japan	Hashizume 2022^Citation100	2012–2018	≥40	All cases	NR	NR	7.1%^d	NR	NR
Japan	Ishikawa 2022^Citation74	2016–2018	All ages	Hospitalized HZ	NR	NR	N/A	Median: 9	NR
Korea	Cheong 2014^Citation101	2009^b	≥50	All cases	NR	4.0	4.8%^e,f	10.9	0.9%^g
Korea	Choi 2014^Citation102	2007–2010	All ages	Immunocompetent	NR	Median: 3.0	36.6%^e,f	Median: 7.8	NR
Korea	Kim 2014^Citation30	2011	All ages	All cases	NR	NR	• All ages: 0.05 • ≥50 YOA: 6.2%	NR	NR
Korea	Song 2014^Citation103	2009–2010	≥50	All cases	7.0	5.0	32.5%^d (over 180 days)	8.9	1.3
Korea	Hong 2016^Citation94	2009–2013	All ages	PHN^h	NR	1.0	• Hospital visits: 4.7 • General hospital visits: 3.8	NR	NR
Korea	Kim 2019^Citation43	2002–2013	≥20	All cases	NR	NR	7.7%^d	NR	NR
Taiwan	Jih 2009^Citation47	2000–2006	All ages	All cases	NR	NR	NR	8.3^f	NR
Taiwan	Lin 2010^Citation49	2000–2005	All ages	All cases	NR	NR	2.9%^f	8.3	NR
Taiwan	Tsai 2015^Citation83	2008–2009	≥50	All cases	6.0	2.8	20.7%^d (over 180 days)	NR	1.3
Taiwan	Lu 2018^Citation46	2004–2008	All ages	All cases	NR	25.6	0.24^f	NR	0.42
				Non-PHN	NR	25.5	0.24^f	NR	0.42
				PHN	NR	30.4	0.28^f	NR	0.54

^aData are presented for all HZ-related hospitalizations (principal and non-principal HZ diagnoses), unless otherwise stated. Data for the proportion of HZ cases hospitalized are within the study population over the study period; ^bAnnualized data presented; ^cExcess HCRU compared with matched control; ^dStudy did not specify if hospitalization data included patients with principal and/or non-principal HZ diagnosis; ^eProportion of patients with HZ within the study area, not necessarily enrolled in the prospective part of the study; ^fHospitalization for principal HZ diagnosis only; ^gPercentage of those with any history of an emergency department visit among all HZ patients; ^hPatients with primary or secondary PHN diagnoses. ED: emergency department; GP: general practitioner; HCRU: healthcare resource utilization; HZ: herpes zoster; LOS: length of stay; N/A: not applicable; NR: not reported; PHN: postherpetic neuralgia; w/: with; w/o: without; YOA: years of age.

The proportion of HZ cases hospitalized among older adults were 3.3% in Australia (≥45 YOA),²¹ 1.9–7.1% in Japan (≥40 YOA),^26,69,100 4.8–6.2% in Korea (≥50 YOA),^30,101 and 0.4–6.9% in China (≥50 YOA).^53,54,59 Studies in Taiwan with data for different age subgroups reported that the proportion of hospitalized HZ cases increased with age, from 1.8% to 6.6% (50–64 YOA to ≥ 85 YOA)⁴⁶ and from 4.1% to 10.3% (60–69 YOA to ≥ 80 YOA).⁴⁹ One study each in Korea and Taiwan reported high proportions of hospitalization among HZ cases aged ≥ 50 YOA (32.5%¹⁰³ and 20.7%,⁸³ respectively); the vast majority of patients enrolled in these two studies experienced moderate to severe pain at enrollment, suggesting that these study populations represented patients with greater HZ severity. There was a higher proportion of males with HZ who required hospitalization than females in China and Japan.^26,53 In Australia, the proportion of patients (≥45 YOA) with a first HZ episode and recurrent HZ episode who were hospitalized were comparable (2.5% and 2.7%, respectively).⁵⁰

In Australia, specialist referrals were observed in 1.8% of encounters for HZ or PHN in patients ≥ 50 YOA, and this proportion of referrals increased with age.⁵¹ In China, a survey among physicians showed that the number of all-cause and PHN-related ED visits per year were the same in patients aged ≥40 years, whereas more than half of all-cause outpatient visits and hospitalizations were PHN-related.⁹⁹ Total all-cause HCRU costs were driven by hospitalizations, whereas PHN-related out-of-pocket costs were driven by the cost of prescription PHN medication.⁹⁹ In a retrospective survey among HZ patients aged ≥50 years, outpatient and inpatient costs were observed to increase in cases with sequelae, and with increasing age of HZ onset.⁵³ Database analyses from 2015–2017 found that patients with complications had more frequent outpatient visits than those without complications (average 5.6 vs 3.3 visits)⁵⁴; however, outpatient medical costs were higher (about 1.2 times) for patients without than with complications, the reasons for which were unclear.⁵⁴

HZ-related costs

Generally, HZ-related costs were higher in study populations of older adults versus overall population (all ages) and in HZ cases with PHN or other complications; Table 3 summarizes findings related to costs and the cost components used in the various studies.

Table 3. HZ-related adjusted-direct and indirect costs, and cost components across selected locales in Asia-Pacific.

					Direct cost
Locale	Source	Study period	Age (years)	Patient disease profile	Total all-cause cost (USD)	Cost components	Indirect cost (USD)
Mean per HZ case
Australia	Stein 2009^Citation51	General practice (2000–2006) and hospitals (1998–2005)	≥50	All cases	832.78–870.00	Healthcare resource (GP visits, outpatient visits, and ED visits, hospitalization)	NR
China	Li 2016^Citation53	2010–2012 (costs converted into 2013 price)	≥50	All cases	187.51^a	Drugs (OTC), healthcare resource (outpatient visits, hospitalization), others (transport, productivity loss [caring for patient], cost considered to be associated with HZ)	NR
				w/sequelae	272.56^a
				w/o sequelae	118.31^a
China	Yu 2019^Citation99	2016	≥40	w/PHN	Annual: 2,062.89	Drugs (TCM, prescription medicine), therapy (physical therapy, acupuncture, massage, heat therapy, exercise, transcutaneous electrical nerve stimulation, chiropractor, yoga, spinal decompression), healthcare resource (consults, ED visits, hospitalization)	Annual PHN-related productivity loss • Absenteeism: 2,437.36 • Presenteeism: 3,746.05 • Total indirect costs: 5,546.15
Hong Kong	Ming 2019^Citation71	2011–2013	≥50	All hospitalized cases	5,734.85	Drugs (medication), healthcare resource (hospitalization), laboratory tests	NR
Japan	Honda 2017^Citation104	2005–2013	≥18	w/PHN	6-month treatment period: • Up to May 2010: 1,337.56 • From Jun 2010: 1,503.78	Drugs (pregabalin, pain-relief treatment, prescription medicine), healthcare resource (outpatient visits, hospitalization), procedures	NR
Japan	Nakamura 2017^Citation69	2013–2015	≥60	All cases	484.26	Drugs (medications, injections, ointment treatment), healthcare resource (outpatient visits, hospitalization), other procedures	HZ-related productivity loss: 119.86
				w/o complications	394.14		NR
				w/PHN	1,056.55
				w/non-PHN complications	745.74
Japan	Imafuku 2019^Citation70	2005–2014 (costs evaluated for 2012–2014)	≥18	All cases	382.16	Drugs (prescription medications [antivirals, analgesics, dermatological], injections, topical, other HZ-related treatments), healthcare resource (outpatient visits, hospitalization), laboratory tests	NR
				w/o complications	287.50
				w/PHN	1,315.57
				w/non-PHN complications	671.06
Korea	Cheong 2014^Citation101	2009 (costs inflated to 2013 price)	≥50	All cases	• 264.69 (HZ management) • 254.17 (PHN management)	Drugs (antivirals, non-narcotics, anti-epileptics) and amount reimbursed by the NHI to medical providers	NR
				Immunocompetent	• 242.13 (HZ management) • 228.60 (PHN management)
Korea	Choi 2014^Citation102	2007–2010	All ages	Immunocompetent	Median: 582.74	Drugs (medication), healthcare resource (inpatient, outpatient), procedures	NR
Singapore	Chen 2015^Citation75	2010–2013	≥50	All cases	370.89^a	Drugs (antiviral, pain medicine), healthcare resource (consultation, hospitalization)	NR
				w/PHN	392.14		Absenteeism: 60.23
				w/o PHN	203.66		Absenteeism: 87.88
Taiwan	Jih 2009^Citation47	2000–2006	All ages	All cases under home care	133.98	Drugs (NSAIDS, acetaminophen, antivirals, systemic corticosteroid, others [anti-depressants, anti-convulsants, opiates, Chinese herbal medicine, topical NSAID, acupuncture, lidocaine cream]), healthcare resource (outpatient visits, hospitalization)	NR
				All hospitalized cases	3,080.36
Annual socioeconomic cost
Korea	Choi 2010^Citation44	2003–2007	All ages	All cases	• 2003: 191.3 mil • 2004: 221.8 mil • 2005: 244.2 mil • 2006: 262.6 mil • 2007: 276.7 mil	Drugs (medication), healthcare resource (clinic visits, hospitalizations), others (transport, nursing)	HZ-related productivity loss • 2003: 30.4 mil • 2004: 33.1 mil • 2005: 37.6 mil • 2006: 41.7 mil • 2007: 44.8 mil
Taiwan	Lin 2010^Citation49	2000–2005	All ages	All cases	• 2000: 25.8 mil • 2004: 27.5 mil	Medical expenditure related to HZ treatment (unspecified)	NR

Only studies that reported total all-cause direct cost data and corresponding cost components are included. To facilitate comparisons between different locales and to account for inflation, all HZ-related cost estimates were adjusted to 2022 values based on the worldwide inflation rate from the World Bank Database, then converted to United States Dollars (USD) if the cost was reported in another currency, based on 2022 currency conversion rates obtained from the United Nations Conference on Trade and Development (UNCTAD) database. Therefore, numerical values may differ from those reported within the referenced articles. ^aInclusive of direct costs and indirect costs from productivity loss. ED: emergency department; GP: general practitioner; HZ: herpes zoster; NHI: National Health Insurance; NR: not reported; NSAID: nonsteroidal anti-inflammatory drug; OTC: over the counter; PHN: postherpetic neuralgia; TCM: traditional Chinese medicine; UNCTAD: United Nations Conference on Trade and Development; USD: United States Dollar; w/: with; w/o: without.

Total all-cause direct medical cost per HZ case varied by geographical regions, at approximately USD 833–870 in Australia,⁵¹ USD 188 in China,⁵³ USD 382–484 in Japan,^69,70 USD 265 in Korea,¹⁰¹ and USD 371 in Singapore⁷⁵ (all costs converted to 2022-adjusted USD to facilitate comparison; Table 3). The direct medical cost per hospitalized HZ case was around USD 5,735 in Hong Kong⁷¹ and USD 3,080 in Taiwan.⁴⁷ The components of direct medical cost measures varied from medications only or medical visits/hospitalization only, to inclusion of all medications, outpatient/hospital utilization, and procedures combined (Table 3). Overall, older people accounted for a higher proportion of HZ-related medical expenditure, for example, patients aged ≥60 years accounted for 59.5% of the total medical care costs in Taiwan.⁴⁷ Costs were also higher for HZ patients with PHN or other HZ-related complications than those without complications (Table 3).^53,69,70,75

Indirect cost from productivity loss, although varying in definition between studies, was shown to pose an important economic burden to patients and caregivers (Table 3). In China, work impairment among employed individuals resulted in an annual indirect cost of USD 5,546.⁹⁹ These patients self-reported in a questionnaire that PHN resulted in a loss of 26.2% in productive time through lost days from work (absenteeism) and 48.4% in productive time while at work (presenteeism). There was an overall work impairment of 55.6% in employed individuals, and an overall activity impairment other than work of 51.1% in all respondents regardless of employment status.⁹⁹ In Singapore, indirect cost due to absenteeism at work was the highest in the labor-productive age group of 50–59 years.⁷⁵ In Japan, indirect cost due to absenteeism in patients and caregivers increased with age owing to higher productivity loss among caregivers of older patients, and represented 19% of total HZ-related costs from the societal perspective.⁶⁹

The estimated socioeconomic burden for HZ-related direct medical cost was reported to be approximately USD 25.8–27.5 million in Taiwan⁴⁹ and USD 191.3–276.7 million in Korea⁴⁴ (all costs converted to 2022-adjusted USD to facilitate comparison; Table 3). Total cost of HZ was also shown to increase over time. In Taiwan, total direct cost increased to approximately 1.1 times (adjusted for inflation) from 2000 to 2004.⁴⁹ In Korea, the total socioeconomic cost (including both direct and indirect costs) was USD 221.7–321.4 million per year in 2003–2007, increasing every year by 14–20%, where both direct and indirect cost increased year-by-year⁴⁴; a separate study reported that in 2009–2013, the cost of PHN management increased by around 40%, where this corresponded with an approximately 58% increase in the number of PHN cases in 2013 versus 2009.⁹⁴

Quality of life

Four studies reporting QoL-related measures were identified.^83,96,99,103 The EuroQol 5-dimensions (EQ-5D) utility score of patients decreased after rash onset compared with scores prior to HZ.^83,103 EQ-5D utility scores subsequently improved in HZ patients over the study follow-up periods,^83,96,103 but not in patients with PHN.⁹⁶ HZ-related QoL findings are summarized in Table 4.

Table 4. HZ-related quality of life outcomes across selected locales in Asia-Pacific.

Locale (Source)	Setting	QoL outcomes
China (Yu 2019)^Citation99	Hospital-based study, 2016, PHN patients ≥40 years	Patient-reported outcomes collected through EQ-5D-3 L questionnaire. Mean (SD) EQ-5D-3 L scores of 0.65 (0.20) on the calculated health utility index and 60.1 (18.6) on the visual analog scale at the time of survey completion.^a Self-reported pain of moderate severity and mild-to-moderate pain interference, poor HRQoL, short duration and poor quality of sleep.
Japan (Mizukami 2018)^Citation96	Hospital/clinic-based study, 2013–2015, ≥60 years	Patient-reported outcomes collected through EQ-5D and SF-12 questionnaires. Pain score and impaired QoL were almost resolved by Day 90 from the onset of HZ rash (Day 0) only in patients who did not develop PHN, but not in patients with PHN. Mean (SD) EQ-5D score increased from 0.76 (0.14) on Day 0 to 0.95 (0.14) on Day 90 in HZ patients without PHN. Mean (SD) EQ-5D score was comparable at 0.68 (0.14) on Day 0 and 0.75 (0.13) on Day 90 in HZ patients with PHN.
Korea (Song 2014)^Citation103	Hospital-based study, 2009–2010, HZ cases ≥50 years	Patient-reported outcomes collected through EQ-5D questionnaire. Mean (SD) EQ-5D score decreased from 0.91 (0.12) prior to HZ to 0.65 (0.16) after HZ onset. QoL subsequently improved, however EQ-5D score remained significantly lower until 180 days of follow-up, compared with before the onset of HZ.
Taiwan (Tsai 2015)^Citation83	Hospital/clinic-based study, 2008–2009, ≥50 years	Patient-reported outcomes collected through EQ-5D questionnaire. Mean (SD) EQ-5D score significantly decreased from 0.91 (0.16) before rash onset to 0.67 (0.18) after rash onset at study enrollment, showing significant QoL deterioration up to 60 days. Mean (SD) EQ-5D increased slightly from 0.67 (0.18) after rash onset at study enrollment to 0.85 (0.18) at 30 days of follow-up. Impact of HZ on QoL and pain severity was similar across age groups.

^aPatients identified with PHN during a specialist office visit were recruited to complete the questionnaire and the time relative to initial HZ diagnosis was not reported. EQ-5D: EuroQol 5-dimensions; EQ-5D-3 L: EuroQol 5-dimensions 3-level; HRQoL: health-related quality of life; HZ: herpes zoster; PHN: postherpetic neuralgia; QoL: quality of life; SD: standard deviation; SF-12: Short-Form 12 version 2.0.

Discussion

This SLR consolidates the latest data on HZ epidemiology and burden among the general population in selected Asia-Pacific locales; the observed population trends support and supplement existing literature. HZ incidence rates observed in this study (0.3–10.4/1,000 PY) were largely comparable to those previously reported in broader Asia-Pacific populations (3–10/1,000 PY).⁶ The one-third (~30%) lifetime risk of HZ reported in Japan²⁵ and Taiwan⁴⁹ is similar to that observed in other populations.^2,105 Age was a predominant factor influencing HZ incidence, similar to previous observations.^6,14,15 HZ incidence rates in the general population ≥ 50 YOA (3.4–16.7/1,000 PY) were consistent with reports elsewhere (3.4–19.5/1,000 PY)^14,15 and tended to be similar or higher compared with published estimates in Europe and North America (5.2–10.9/1,000 PY and 6.6–9.0/1,000 PY, respectively).^14,15 The slight decline observed in HZ incidence rates after 70 YOA in our study, while differing from observations in the United States and Europe that show increasing HZ incidence with increasing age,^105–108 is generally consistent with previous reports in Asia-Pacific populations where HZ incidence declined in the highest age groups,^6,15,108 although the reasons for which have not been fully explored and may be areas for future research. Specifically, it would be important to ascertain if the observed geographical differences (APAC versus non-APAC or amongst APAC locales) are due to underlying biological factors, or if the observations reflect differences in local culture, access to healthcare, and healthcare utilization.

HZ incidence rates were generally higher in females compared with males in the Asia-Pacific region, congruent with published APAC and global reports.^{6,14,15,106,109} While this trend remains unexplained in the literature, a recent United States study suggested that factors other than hormonal differences and variations in health-seeking behaviors need to be considered when elucidating the association between HZ and sex.¹¹⁰ In general, HZ incidence rates increased over time, based on evidence from locales with available data for multiple consecutive years,^{24,33,41,42,46} consistent with trends observed globally.¹⁵ Literature gaps regarding the impact of HZ vaccination on HZ incidence were observed.

The incidence rates of HZ varied among the locales in the Asia-Pacific region included in this SLR. This heterogeneity was also observed in previous studies^6,14,15 and could possibly be due to differences in healthcare systems, healthcare seeking behaviors, data sources (medical records, surveys, insurance claims data, and prescription databases), definition of HZ cases (International Classification of Disease [ICD] codes, clinical exam, use of antiviral prescriptions, and self-reports), study population compositions, and study periods.^107,111,112 HZ incidence rates reported in Korea were higher than other locales included in this SLR, consistent with reports elsewhere.^6,14 This has been suggested to be due to environmental and cultural factors in Korea, whereby access to state-insured healthcare and high public awareness of HZ likely result in patients seeking medical attention even if their symptoms are mild.³⁰ In China, community surveys among the general population tended to report lower HZ incidence compared with other locales in this SLR.^52,53,59 Incidence estimates from these survey-based studies were previously also identified as outliers in a meta-analysis¹⁴; the data may have been impacted by recall bias, where mild to moderate infection were likely to be underreported by patients, and the severe cases reported were likely to develop recurrence. In contrast, HZ incidence reported from a database study in China was more comparable to the findings from other locales.⁵⁴ Overall, the variability of epidemiological estimates of HZ and HZ complications across studies underscore the need for standardized methodology to better estimate the disease burden and assess the impact of HZ vaccination in APAC populations.

The rates of HZ-related hospitalization increased with age and in persons with PHN or other complications. HCRU and costs varied widely given the various study designs, cost components included in each study, and different healthcare systems and practices across locales. Nonetheless, HCRU was generally found to also be higher in patients of older age and in the presence of PHN or other complications, and annual socioeconomic costs increased over time. These data suggest that timely HZ prevention in adults and populations vulnerable to HZ-related complications may be advantageous in reducing the overall burden of HZ in the Asia-Pacific region and warrants further evaluation.

Overall, the population trends of HZ epidemiology and burden from the current SLR reinforces the increasing burden of HZ and need for preventive strategies in the Asia-Pacific region, especially with its aging population that would live through a greater number of years at risk of HZ and HZ-related complications. There is an urgent and important need to educate relevant stakeholders on the burden of HZ and the availability of preventive measures, as well as encourage timely uptake of HZ vaccination prior to occurrence of peak disease incidence to relieve the overall burden of HZ on healthcare systems in the region. Notably, the durability of protection of vaccines against HZ is a crucial consideration to long-term public health policies and cost impacts of HZ vaccination. Recent studies have reported efficacy and real-world effectiveness of RZV against HZ (81.6% up to 10 years post-vaccination and 76% up to 4 years post-vaccination in fully vaccinated individuals, respectively).^113,114 While further long-term data are necessary, these available data suggest it may be beneficial to target adults at an earlier age than that of those most likely to receive HZ vaccination currently. The impact of varicella vaccination on HZ epidemiology was beyond the scope of the current SLR. Nevertheless, it should be noted that studies in the United States have found that childhood varicella vaccination was associated with lower HZ risk in children^110,115,116; whether HZ prevention strategies in the Asia-Pacific region should consider varicella vaccination among children and susceptible adults warrants separate investigation.

The findings of the current study are consistent with existing literature that call for improved HZ preventive strategies in the Asia-Pacific region.^6,10,13,109 Chen et al.⁶ previously observed that local adult HZ immunization guidelines exist in some Asia-Pacific locales. However, those guidelines were often underused due in part to the lack of local evidence available and understanding of the local HZ epidemiology and burden at the time of that SLR (December 2014).⁶ Despite the increased availability of HZ vaccines since then, a recent review of HZ vaccine guidelines found that as of 2022, formal recommendations regarding HZ vaccination were still few globally, including in the Asia-Pacific region.¹⁰ These observations underscore the importance of consolidating rapidly evolving epidemiological evidence and the projected disease burden of HZ in Asia-Pacific locales, to guide the formulation of optimal immunization strategies and guidelines (e.g., age-based vaccination recommendations). Moreover, the trends identified for each locale in this SLR can help to inform the direction of future studies, with potential impact on local vaccination recommendation policies (e.g., optimizing vaccine accessibility among sub-populations based on projected lifetime risk and public health impact of vaccination).

Finally, despite these findings being consistent with previous research, this SLR has some limitations that need to be considered in the overall interpretation of the results. Firstly, age-related analysis was limited by the availability of age-related data due to inherent differences across studies. For example, some studies provided data on HZ incidence and hospitalization for a broad age range (e.g., all ages, ≥50 YOA only), while other studies provided data across smaller age bands (e.g., 40–49 YOA, 50–59 YOA), which were not directly comparable. Similarly, for data related to the study period, some studies reported outcomes for a specific year, while other studies reported outcomes as an average over time (up to a decade) without year-specific data, thereby limiting any year-related and time trend analyses. Secondly, the definitions of HZ and PHN were heterogeneous across studies; differences in case definitions have been reported to affect estimates of epidemiology, HCRU, and cost-related data, meaning any generalized comparisons should be interpreted with caution.^{107,112,117,118} Next, cost data were adjusted based on local inflation rates then exchanged to USD. While using local inflation rates more accurately reflects the price changes for local healthcare resources compared with using US inflation rates, there are inherent limitations to this method that may result in overestimated adjusted costs.¹¹⁹ Moreover, there were none-to-limited data in certain locales pertaining to HZ recurrence (Hong Kong, New Zealand, Singapore, and Taiwan), HCRU (New Zealand and Singapore) and QoL or other patient reported outcomes data (Australia, Hong Kong, New Zealand, and Singapore). Sex-specific data were also not always available. Furthermore, the SLR search was limited to articles published in English and on indexed databases only, presenting a potential gap in the literature search. Future reviews may benefit from including articles in local languages and gray literature (e.g., government reports and statistics) for a more inclusive and comprehensive overview of the burden of HZ across Asia-Pacific.

In conclusion, the healthcare and economic burden of HZ and its complications are expected to increase as the aging population in the Asia-Pacific region grows. Decision-making on HZ vaccine coverage can be informed by health technology assessments that consider the following aspects: firstly, the projected risk or burden of HZ on the aging population including local trends; secondly, the populations that are most vulnerable to HZ in the local population; and thirdly, existing or planned initiatives designed to raise awareness and provide education on HZ. Finally, as real-world evidence on HZ epidemiology and burden continuously evolves, evidence generation strategies must concurrently progress and adapt, so that HZ-related policies can be shaped based on the most up-to-date and locally relevant evidence.

Acknowledgments

The authors acknowledge Chia Jie Tan, PhD, for initial literature screening, and Roeland Van Kerckhoven, PhD, GSK for publication coordination. The authors also thank Costello Medical for editorial assistance and publication coordination, on behalf of GSK, and acknowledge Kyra Chan, PhD, and Sharon Lee, PhD, Costello Medical, Singapore, for medical writing and editorial assistance based on the authors’ input and direction.

Author’s contributions

JC and SS conceived this systematic review. PA performed the literature search. PA, YK, JC, and CRO identified the eligible studies and analyzed the data. All authors participated in the interpretation of the study and the development of this manuscript. All authors had full access to the data and gave final approval before submission.

Disclosure statement

JC, RP, and SS are employed by and have stock ownership in the GSK Group of Companies. YK and CRO were employed by the GSK Group of Companies at the time of the study. PA reports consulting services to the GSK Group of Companies and Merck Sharp & Dohme Corp.

Data availability statement

This systematic literature review protocol was not registered. Data sharing of anonymized subject level data is not applicable to this article. All data supporting the findings of this study were obtained from the references cited.

Additional information

Funding

GlaxoSmithKline Biologicals SA funded this study (GSK study identifier: VEO-000196) and was involved in all stages of study conduct, including analysis of the data. GlaxoSmithKline Biologicals SA also covered all costs associated with the development and publication of this manuscript, including support for third-party writing assistance for this article in accordance with Good Publication Practice (GPP 2022) guidelines (https://www.ismpp.org/gpp-2022).