Cost-utility and cost-benefit analysis of pediatric PCV programs in Egypt

Figures & data

Figure 1. (a) Markov cycle tree of Streptococcus pneumoniae. The structure of the “Program B” branch is the same as that of the “Program A” branch. SP: Streptococcus pneumoniae; OCAP: Outpatient pneumococcal pneumonia; ICAP: Inpatient pneumococcal pneumonia; AOM: Pneumococcal acute otitis media; OCAP_TD: Temporary disability after OCAP; ICAP_TD: Temporary disability after ICAP; MEN_TD: Temporary disability after meningitis; BACT_TD: Temporary disability after bacteremia; AOM_TD: Temporary disability after AOM; NDI: Neurodevelopmental impairment after meningitis. (b) Cycle tree for at least one major sequela from inpatient pneumonia. The structures of nodes C, D, E, and F in Figure 1a are the same as that of node B in Figure 1b. (c) Cycle tree for the multiple sequelae state.

Figure 1. (Continued).

Table 1. Markov model input parameters.

Base-case (most unfavorable case, most favorable case)
Parameter:	0–1 years	2–4 years	5–17 years		18–34 years	35–49 years	50–64 years	65+ years	Data source
Incidence per 100,000
SP Pneumonia (outpatient)^a	523 (489, 637)	523 (489, 637)	216.14 (202, 263)		104.72 (98, 128)	121.31 (113, 148)	247.36 (231, 301)	681.97 (638, 831)	a
SP Pneumonia (inpatient)^b	334 (250, 381)	334 (250, 381)	138.03 (103, 157)		66.88 (50, 76)	77.47 (58, 88)	157.97 (118, 180)	435.52 (326, 497)	b
SP AOM^c	9,499.7 (6410, 13,555)	10,279 (6787, 15,033)	2,593.98 (1716, 3893)		0	0	0	0	c
IPD^d	20 (6, 47)	20 (6, 47)	4.88 (1.46, 11.46)		2.47 (.74, 5.8)	1.64 (.49, 3.86)	1.5 (.45, 3.53)	1.97 (.59, 4.62)	d
Meningitis percent of IPD	34.84% (not varied)	13.11% (not varied)	9.52% (not varied)		8.96% (not varied)	8.82% (not varied)	7.14% (not varied)	2.39% (not varied)	Wilson et al. (2018)^Citation22
Case fatality rates
Hospitalized pneumonia^e	.068 (.055, .082)	.068 (.055, .082)	.005 (.004, .006)		.012 (.01, .014)	.029 (.023, .035)	.043 (.034, .051)	.107 (.086, .129)	e
Meningitis^f	.568 (.454, .681)	.568 (.454, .681)	.17 (.136, .205)		.17 (.136, .205)	.187 (.15, .225)	.194 (.155, .233)	.406 (.324, .487)	f
Bacteremia and sepsis^g	.392 (.314, .47)	.392 (.314, .47)	.392 (.314, .47)		.223 (.179, .268)	.223 (.179, .268)	.223 (.179, .268)	.223 (.179, .268)	g
Baseline health utility	.97 (not varied)	.97 (not varied)	.94 (not varied)		.92 (not varied)	.89 (not varied)	.88 (not varied)	.82 (not varied)	Chuck et al. (2010)^Citation36
	All ages	Data source
Health utilities
Meningitis	.9638 (.973, .9546)	Kulpeng et al. (2013)^Citation21
Bacteremia and sepsis	.9852 (.9902, .9802)	Kulpeng et al. (2013)^Citation21
AOM	.9984 (.9986, .9982)	Kulpeng et al. (2013)^Citation21
Pneumonia (outpatient)	.994 (.9999, .964)	Aljunid et al. (2014)^Citation37
Pneumonia (inpatient)	.992 (.9999, .962)	Aljunid et al. (2014)^Citation37
Hearing loss after AOM	.9984 (.9986, .9982)	Assumed to equal the utility weight of AOM
Epilepsy after meningitis	.64 (.7876, .4924)	Kulpeng et al. (2013)^Citation21
Hearing loss after meningitis	.55 (.6608, .4392)	Kulpeng et al. (2013)^Citation21
NDI after meningitis	.1 (.317, .05)	Kulpeng et al. (2013)^Citation21
Long-term sequelae after pneumonia	.59 (.705, .475)	Kulpeng et al. (2013)^Citation21
Multiple sequelae	.37 (.444, .296)	Sibak et al. (2015)^Citation19
Long-term sequelae probabilities
Epilepsy after meningitis	.103 (not varied)	Netsawang et al. (2010)^Citation38
Hearing loss after meningitis	.034 (not varied)	Netsawang et al. (2010)^Citation38
NDI after meningitis	.345 (not varied)	Netsawang et al. (2010)^Citation38
Hearing loss after AOM	.051 (.0498, .0522)	MoPH, Thailand; 2007^Citation39
Sequelae after pneumonia	.186 (.15, .234)	Edmond et al. (2012)^Citation40
Direct medical costs per-care episode (2016 USD)h
AOM	1.816 (1.569, 2.119)	Sibak et al. (2015)^Citation19
Pneumonia (outpatient)	2.785 (2.612, 3.53)	Sibak et al. (2015)^Citation19
Pneumonia (inpatient)	31.414 (28.262, 33.836)	Sibak et al. (2015)^Citation19
Meningitis	240.415 (201.33, 245.89)	Sibak et al. (2015)^Citation19
Bacteremia	96.834 (87.24, 106.92)	Sibak et al. (2015)^Citation19

Abbreviations: SP = Streptococcus pneumoniae; AOM = acute otitis media; IPD = invasive pneumococcal disease; NDI = Neurodevelopmental impairment.

^aConstructed from Wahl et al. (2018)³ Egypt-specific Sp. outpatient pneumonia estimates and 2016 GBD IHME⁴¹ lower respiratory infection incidence rate estimates for Egypt.

^bConstructed from Wahl et al. (2018)³ Egypt-specific Sp. inpatient pneumonia estimates and 2016 GBD IHME⁴¹ lower respiratory infection incidence rate estimates for Egypt.

^cAge-specific 2016 GBD IHME⁴¹ all-cause AOM estimates for Egypt converted to pneumococcal AOM using 26.4% pneumococcal derived from Ngo et al. (2016).⁴²

^dConstructed from Wahl et al. (2018)³ Egypt-specific Sp. meningitis and Sp. bacteremia estimates and 2016 GBD IHME⁴¹ pneumococcal meningitis incidence rate estimates for Egypt.

^eEgypt-specific Wahl et al. (2018)³ estimates of the number of cases and deaths from inpatient Sp. pneumonia for children extrapolated to older ages using age structures from Wu et al. (2016).⁴³

^fConstructed by multiplying the author-constructed inpatient pneumonia CFRs by the age-specific ratio of the Wu et al. (2016)⁴³ meningitis CFR to that for inpatient pneumococcal pneumonia.

^gConstructed by multiplying the author-constructed inpatient pneumonia CFRs by the age-specific ratio of the Wu et al. (2016)⁴³ bacteremia CFR to that for inpatient pneumococcal pneumonia.

^hMedical costs are adjusted for the percent seeking care, inflation, and market exchange rates.

Figure 2. Incidence rate projection model results. Overall invasive pneumococcal disease (IPD) incidence rate projections. Projections under a PCV10 program (red); Projections under a PCV13 program PCV13 (blue). Horizontal axis is time (in years) since the introduction of a universal pediatric PCV program.

Figure 3. Incidence rate projection model results. Effect of pediatric PCVs against invasive pneumococcal disease. PCV effect = ((no program incidence – program incidence)/no program incidence)*100. PCV10 effects (red); PCV13 effects (blue). Population: 2016 and 2023 birth cohorts. Base-case analysis.

Table 2. Health-augmented lifecycle model input parameters.

Values represent the arithmetic mean over the age group
	Age group (years)
Parameter:	0-14	15-24	25-34	35-44	45-54	55-64	65-74	75-84	85-90	91-100
Annual consumption^a (2016 USD)	1,254.04	2,098.62	2,563.73	2,682.35	2,842.78	2,733.84	2,462.50	2,328.45	2,240.25	1,951.60
Annual earnings^b (2016 USD)	0.00	1,248.19	2,920.70	3,727.57	3,848.79	3,182.57	1,911.07	627.76	176.00	0.00
Hourly wage^c (2016 USD)	1.82	1.85	2.06	2.24	2.34	2.33	2.09	1.86	1.82	1.82
Nonmarket time (hours/day)^d	1.40	3.36	4.68	4.90	5.01	7.43	8.98	9.00	9.00	9.00
Average consumer surplus per unit of full consumption^e	0.0037	0.072	0.089	0.093	0.10	0.10	0.10	0.10	0.10	0.10

In the context of the HALM, annual earnings measure the value of market time. Since few individuals younger than age 15 and older than age 90 work, we assign these age groups a value of zero for annual earnings. However, we assign these age groups a non-zero hourly wage since we use hourly wages to monetize the value of nonmarket time.

Data Source: ^aWorld Development Indicators for 2016 total consumption;⁸⁵ United Nations World Population Prospects: 2017 Revision⁸⁶ for 2016 total population; National Transfer Accounts data⁶⁷ for Turkey for age structure.

^bWorld Development Indicators for 2016 per capita GDP;⁸⁰ National Transfer Accounts data⁶⁸ for Turkey for age structure.

^c2016 International Labor Organization data^69,70 for Egypt for the overall average hourly wage; age structure constructed by pooling data from: Office of National Statistics Algeria;⁸⁷ Statistical Committee of the Republic of Armenia;⁸⁸ Republic of Bulgaria National Statistical Institute;⁸⁹ and Eurostat⁹⁰ Turkish data.

^dTurkish Statistical Institute;⁷¹ Office of National Statistics Algeria.⁷²

^eAuthors’ calculations (see Supplemental File A2).

Figure 4. Markov model results. Quality-adjusted life year gains for select birth cohorts. Base-case analysis.

Figure 5. Health-Augmented lifecycle model results. Base-case analysis using the 2016 birth cohort. Value of a statistical disability year (black); Value of a statistical life year (dashed black); Full income (green); Full consumption (red); Annual earnings (blue).

Figure 6. Markov model results. Vaccination effects on lifetime survival probability and lifetime health utility. Base-case analysis using the 2016 and 2023 birth cohorts. PCV10 effects (red); PCV13 effects (blue).

Figure 7. Willingness to pay (WTP) results. WTP for mortality risk reductions and morbidity risk reductions. Base-case analysis using the 2016 and 2023 birth cohorts. WTP for PCV10 (red); WTP for PCV13 (blue).

Table 3. Base-case incremental cases, hospital episodes, and mortality over a 100-year time horizon.

Percentiles are from multivariate probabilistic sensitivity analysis
	Program			Incremental
	No program	PCV10	PCV13	PCV10 versus no program	PCV13 versus no program	PCV13 versus PCV10
Cases
IPD	598,144	592,986	315,685	−5,158	−282,458	−277,301
Pneumonia	61,910,259	59,242,670	56,357,710	−2,667,589	−5,552,549	−2,884,960
AOM	197,103,639	165,769,521	132,902,189	−31,334,118	−64,201,451	−32,867,333
Total cases (TC)	259,612,042	225,605,177	189,575,584	−34,006,864	−70,036,458	−36,029,593
2.5^th percentile TC	184,478,341	157,491,016	127,267,087	−26,987,325	−57,211,254	−30,223,929
97.5^th percentile TC	364,618,754	325,384,819	284,523,954	−39,233,935	−80,094,800	−40,860,865
Hospital episodes
IPD	469,152	463,418	246,301	−5,734	−222,851	−217,117
Pneumonia	5,790,813	5,541,298	5,271,452	−249,514	−519,361	−269,846
Total hospitalizations (TH)	6,259,965	6,004,716	5,517,752	−255,248	−742,212	−486,964
2.5^th percentile TH	4,287,160	4,149,015	3,919,593	−138,146	−367,567	−229,422
97.5^th percentile TH	8,041,588	7,659,748	6,823,866	−381,840	−1,217,722	−835,882
Mortality
IPD	204,057	197,337	102,523	−6,720	−101,534	−94,814
Pneumonia	1,331,659	1,260,785	1,184,135	−70,874	−147,523	−76,649
Total deaths (TD)	1,535,715	1,458,122	1,286,658	−77,594	−249,057	−171,464
2.5^th percentile TD	846,374	804,541	733,690	−41,833	−112,684	−70,850
97.5^th percentile TD	2,398,297	2,254,634	1,869,162	−143,664	−529,136	−385,472

Direct and indirect effects of IPD allowed for all ages. Direct and indirect effects of non-IPD allowed for ages 0–4 years only.

Table 4. Base-case results (95% confidence interval).

PCV13 vs. no program results per vaccinated infant
QALY gain	0.0462
Vaccination cost (2016 USD)	43.63
Averted direct cost (2016 USD)	0.88
WTP (2016 USD)	251.23
WTP per QALY (2016 USD)	5,439.25
Net benefit (2016 USD)	208.48
ICER (2016 USD)	925.6 (512.08–1,734.9)
Rate of return	487.65% (188.23%-992.86%)
PCV10 vs. no program results per vaccinated infant
QALY gain	0.0192
Vaccination cost (2016 USD)	38.43
Averted direct cost (2016 USD)	0.38
WTP (2016 USD)	100.40
WTP per QALY (2016 USD)	5,235.72
Net benefit (2016 USD)	62.35
ICER (2016 USD)	1,984.41 (1,186.32–3,804.67)
Rate of return	163.84% (32.61%-335.81%)
PCV13 vs. PCV10 results per vaccinated infant
QALY gain	0.0270
Vaccination cost (2016 USD)	5.198
Averted direct cost (2016 USD)	0.498
WTP (2016 USD)	150.83
Net benefit (2016 USD)	146.13
Net costs (2016 USD)	4.70
ICER (2016 USD)	173.98 (87.59–331.23)
Rate of return	3109.37% (1409.91%-6601.6%)

Direct and indirect effects of IPD allowed for all ages. Direct and indirect effects of non-IPD allowed for ages 0–4 years only. 95% confidence intervals for ICERs and RoRs are from multivariate probabilistic sensitivity analysis.

Definitions: QALY gain = difference in quality-adjusted life year (QALY) gains between the PCV program and comparator program. Vaccination costs = difference between the PCV program and comparator program in vaccination costs. Averted direct costs = difference between the PCV program and comparator program in direct medical cost. WTP = willingness-to-pay. WTP in PCV13 vs. PCV10 comparison = difference between PCV13 and PCV10 in willingness-to-pay. WTP per QALY = ratio of WTP and QALY gain. Net benefit of a PCV program compared to no program = WTP plus averted direct cost minus vaccination costs. Net benefit of PCV13 compared to PCV10 = difference between PCV13 and PCV10 in net benefits. Net costs in PCV13 vs. PCV10 comparison = difference between PCV13 and PCV10 in vaccination costs minus averted direct costs. ICER = incremental cost-effectiveness ratio, which is the ratio of net costs and QALY gains. Rate of return to a PCV program = ratio of net benefits to net costs (vaccination costs minus averted direct costs) multiplied by 100. Rate of return to PCV13 vs. PCV10 = ratio of the difference in net benefits to the difference in net costs multiplied by 100.

Table 5. One-way deterministic sensitivity analysis.

	Rate of return			ICER (2016 US dollars)
	PCV10 vs. no program	PCV13 vs. no program	PCV13 vs. PCV10	PCV10 vs. no program	PCV13 vs. no program	PCV13 vs. PCV10
Base case	163.84%	487.65%	3109.37%	1,984	926	174
Base-year incidence rates
Most unfavorable case	60.32%	253.26%	1767.88%	3,263	1,504	287
Most favorable case	266.66%	811.07%	5455.50%	1,422	612	105
Implicit vaccine efficacy
Decrease by 20%	111.02%	369.69%	2424.38%	2,481	1,158	221
Increase by 20%	216.68%	605.77%	3818.48%	1,653	771	142
Case fatality rates
Decrease by 20%	143.69%	430.02%	2748.47%	2,167	1,029	196
Increase by 20%	181.98%	540.77%	3445.66%	1,843	847	158
Probability of hearing loss after AOM
Most unfavorable case	161.86%	484.35%	3095.64%	1,997	930	175
Most favorable case	165.87%	491.01%	3123.39%	1,971	921	173
Probability of long-term sequelae after ICAP
Most unfavorable case	141.18%	447.00%	2923.34%	2,158	993	185
Most favorable case	194.06%	541.84%	3357.38%	1,793	848	161
Health utilities
Most unfavorable case	127.67%	420.00%	2786.91%	2,286	1,047	194
Most favorable case	200.56%	554.29%	3418.30%	1,739	825	157
Direct medical costs
Most unfavorable case	163.53%	486.20%	3069.22%	1,987	928	176
Most favorable case	164.21%	489.23%	3152.96%	1,982	923	172
VSL of the average working-age adult
Decrease by 10%	135.44%	424.42%	2764.20%	n/a	n/a	n/a
Increase by 10%	191.59%	549.47%	3447.07%	n/a	n/a	n/a
Discount rate
Change to 0%	459.76%	1149.71%	6758.45%	1,154	530	98
Change to 3.5%	130.09%	412.06%	2693.26%	2,179	1,020	192
Change to 5%	54.10%	241.79%	1756.70%	2,843	1,345	256

Definitions: ICAP=inpatient community-acquired pneumonia; VSL = value of a statistical life. Rate of return equals the net benefits of vaccination as a percent of the net costs of vaccination, where net benefits are equal to the willingness-to-pay plus averted direct medical cost minus vaccination costs and net costs are equal to vaccination costs minus averted direct medical costs. ICER = incremental cost-effectiveness ratio, which is the ratio of net costs and QALY gains.

Most unfavorable and most favorable parameter values are reported in Table 1.

Table 6. Scenario analysis.

	Rate of return			ICER (2016 US dollars)
	PCV10 vs. no program	PCV13 vs. no program	PCV13 vs. PCV10	PCV10 vs. no program	PCV13 vs. no program	PCV13 vs. PCV10
Base case	163.84%	487.65%	3109.37%	1,984	926	174
Allow inpatient pneumonia herd effects for all ages	249.09%	1191.48%	9883.40%	1,677	519	71
Impose cap on overall IPD incidence rate projections	200.71%	488.56%	2802.86%	1,794	925	189
Project vaccine effects using proxy country overall IPD incidence rates	649.20%	779.02%	1769.13%	710	621	318
Reduce study horizon from 100 years to 60 years	67.50%	272.92%	1934.70%	2,554	1,203	228

Definitions: Rate of return equals the net benefits of vaccination as a percent of the net costs of vaccination, where net benefits are equal to the willingness-to-pay plus averted direct medical cost minus vaccination costs and net costs are equal to vaccination costs minus averted direct medical costs. ICER = incremental cost-effectiveness ratio, which is the ratio of net costs and QALY gains.

Wilson M, Wasserman M, Jadavi T, Postma M, Breton M, Peloquin F, Earnshaw S, McDade C, Sings H, Farkough R. Clinical and economic impact of a potential switch from 13-valent to 10-valent pneumococcal conjugate infant vaccination in Canada. Infect Dis Ther. 2018;7(3):353–371. doi:10.1007/s40121-018-0206-1.

 PubMed Web of Science ®Google Scholar

Chuck AW, Jacobs P, Tyrrell G, Kellner JD. Pharmacoeconomic evaluation of 10- and 13-valent pneumococcal conjugate vaccines. Vaccine. 2010;28(33):5485–5490. doi:10.1016/j.vaccine.2010.05.058.

 PubMed Web of Science ®Google Scholar

Kulpeng W, Leelahavarong P, Rattanavipapong W, Sornsrivichai V, Baggett HC, Meeyai A, Punpanich W, Teerawattananon Y. Cost-Utility analysis of 10- and 13-valent pneumococcal conjugate vaccines: protection at what price in the Thai context? Vaccine. 2013;31(26):2839–2847. doi:10.1016/j.vaccine.2013.03.047.

 PubMed Web of Science ®Google Scholar

Aljunid S, Maimaiti N, Ahmed Z, Nur AM, Isa ZM, Azmi S, Sulong S. Economic impact of Pneumococcal Protein-D Conjugate Vaccine (PHiD-CV) on the Malaysian national immunization programme. Value in Health Regional Issues. 2014;3:146–155. doi:10.1016/j.vhri.2014.04.008.

 PubMedGoogle Scholar

Sibak M, Moussa I, El-Tantawy N, Badr S, Chaudhri I, Allam E, Munier A, Abo Freikha S, Hoestlandt C, Lara C. Cost-Effectiveness analysis of the introduction of the pneumococcal conjugate vaccine (PCV-13) in the Egyptian national immunization program, 2013. Vaccine. 2015;33(Suppl 1):A182–A191. doi:10.1016/J.VACCINE.2014.12.044.

 PubMedGoogle Scholar

Netsawang S, Punpanich W, Treeratweeraphong V, Chotpitayasunondh T. Invasive pneumococcal infection in urban Thai children: A 10-year review. J Med Assoc Thai. 2010;93(Suppl. 5):S6–12. PubMed: 21298830

 PubMedGoogle Scholar

Working group of Burden of Disease Project. Nonthaburi: international health policy program. Ministry of public health; Thailand: 2007. Burden of disease and injuries in Thailand; 2004. https://espace.library.uq.edu.au/view/UQ:155542 .

 Google Scholar

Edmond K, Scott S, Korczak V, Ward C, Sanderson C, Theodoratou E, Clark A, Griffiths U, Rudan I, Campbell H Long term sequelae from childhood pneumonia; systematic review and meta-analysis. PLoS ONE. 2012;7(2):e31239. doi:10.1371/journal.pone.0031239.t001.

 PubMed Web of Science ®Google Scholar

Wahl B, O’Brien KL, Greenbaum A, Majumder A, Liu L, Chu Y, Lukšić I, Nair H, McAllister DA, Campbell H, et al. Burden of Streptococcus pneumoniae and Haemophilus influenzae type b disease in children in the era of conjugate vaccines: global, regional, and national estimates for 2000–15. The Lancet Global Health. 2018;6(7):e744–57. doi:10.1016/S2214-109X(18)30247-X.

 PubMed Web of Science ®Google Scholar

Global Burden of Disease. Institute for Health Metrics and Evaluations (IHME); 2016 [accessed 2018 Oct 24]. http://ghdx.healthdata.org/gbd-results-tool .

 Google Scholar

Ngo CC, Massa HM, Thornton RB, Cripps AW, Reid S. Predominant bacteria detected from the middle ear fluid of children experiencing otitis media: a systematic review. PLoS ONE. 2016;11(3):e0150949. doi:10.1371/journal.pone.0150949.

 PubMed Web of Science ®Google Scholar

Wu DB, Roberts C, Lee VWY, Hong L, Tan KK, Mak V, Lee KKC. Cost-Effectiveness analysis of infant universal routine pneumococcal vaccination in Malaysia and Hong Kong. Human Vaccines & Immunotherapeutics. 2016;12(2):403–416. doi:10.1080/21645515.2015.1067351.

 PubMed Web of Science ®Google Scholar

World Bank national accounts data, and OECD National Accounts data files. Series name: “Final consumption expenditure (current US$)”; [accessed 2018 Oct 21].

 Google Scholar

United Nations Population Division. World population prospects: 2017 revision. Series name: “Population, total”; [accessed 2018 Sep 19].https://population.un.org/wpp/Download/Standard/Population/ .

 Google Scholar

National Transfer Accounts; [accessed 2018 Nov 30]. https://ntaccounts.org/web/nta/show/Browse%20database .

 Google Scholar

World Bank national accounts data, and OECD National Accounts data files. Series name: “GDP per capita (current US$)”; [accessed 2018 Oct 18]. https://data.worldbank.org/indicator/PA.NUS.FCRF?end=2016&start=2016 .

 Google Scholar

National Transfer Accounts; [accessed 2017 Apr 16]. http://www.ntaccounts.org/doc/repository/NTA%20profiles.xlsx .

 Google Scholar

International Labour Organization. Data from: ilostat-export.xlsx [dataset]. EGY - EC - Employment, wages and hours of work survey. Mean nominal monthly earnings of employees by sex and economic activity – Harmonized series; 2016 [accessed 2020 May 22]. http://www.ilo.org/ilostat/faces/oracle/webcenter/portalapp/pagehierarchy/Page27.jspx? .

 Google Scholar

International Labour Organization. Data from: iostat-export.xlsx [dataset]. Mean weekly hours actually worked per employed person by sex and economic activity null; 2016 [accessed 2018 Sep 27].

 Google Scholar

Office of National Statistics Algeria. Consumer spending survey and the standard of living of households 2011 report. Tab. n°20: Salaires nets moyens mensuels par sexe et tranches d’âges. Algiers (DZ); 2014 Nov [accessed 2018 Oct 3]. http://www.ons.dz/IMG/pdf/pubfinsal13-2.pdf .

 Google Scholar

Statistical Committee of the Republic of Armenia. Data from: LFS_2016.xlsx [dataset]. Labor force survey anonymised micro data database (by household’s members); 2016 [accessed 2018 Oct 3]. https://www.armstat.am/en/?nid=212 .

 Google Scholar

Republic of Bulgaria, National Statistical Institute. Data from: number of employees, average gross and net hourly earnings and average gross overtime hourly earnings in October 2002 by gender and age - total full-time and part-time employees [dataset]. Time series data on hourly earnings by age; [accessed 2018 Oct 3]. http://www.nsi.bg/en/content/6528/age .

 Google Scholar

European Commission, Eurostat. Data from: mean hourly earnings by sex, age and occupation – NACE Rev. 2, B-S excluding O (earn_ses14_14) [dataset]. Database: the structure of earnings survey; 2014 [accessed 2016 Aug 18]. http://ec.europa.eu/eurostat/data/database .

 Google Scholar

Turkish Statistical Institute (Turkstat). Data from: average activity time per person by type of activity, sex and age group, 2014-2015 [dataset]. Database: Time Use Survey; 2016 [accessed 2016 Nov 2]. http://www.turkstat.gov.tr/PreTablo.do?alt_id=1009.

 Google Scholar

Office of National Statistics Algeria. Enquête Nationale Emploi du Temps ENET 2012. Algiers (DZ); 2013 [accessed 2017 Mar 3]. http://www.ons.dz/

 Google Scholar