Economic burden of Generalized Pustular Psoriasis in Malaysia: a multi-centre analysis of indirect costs Authors: Siew Chin Ong, Xin Qian Seah, Mustapha Mohammed, Mei Ee Tay, Latha Selvarajah, Wooi Chiang Tan, Sook Yee Michelle Voo, Yen Loo Ting, Jyh Jong Tang, Siti Nur Sharida Abd Kadir, Azura Mohd Affandi Published: 2026 ABSTRACT ABSTRACTBackgroundGeneralized Pustular Psoriasis (GPP) constitutes a rare yet severe variant of psoriasis, characterised by extensive pustule formation on the skin and imposes a significant economic burden. The study aimed to investigate the indirect costs associated with GPP in Malaysia, encompassing patients’ work productivity losses and out-of-pocket expenses.MethodsThis primary, multicentre, cross-sectional cost analysis study was conducted among patients diagnosed with GPP attending public hospitals under the Ministry of Health (MOH) in Malaysia. The indirect costs were assessed using the Work Productivity and Activity Impairment (WPAI) tool and standardised data collection form. Data were statistically analysed using descriptive and inferential statistics using IBM SPSS version 28.0.ResultsA total of 54 patients with GPP were enrolled in the study, with about one-fourth experiencing flares at the time of the interview. The total annual indirect cost associated with GPP was estimated at MYR 676,081 (MYR 12,520 per patient). The indirect costs of GPP were driven largely by work productivity loss. Both work and activity impairment were significantly higher during GPP flare periods compared to non-flare periods. Additionally, the annual work impairment cost per patient was substantially higher during GPP flare periods, indicating a marked increase in economic burden compared with the non-flare group.ConclusionThe study demonstrates a substantial indirect economic burden of GPP, driven primarily by work productivity loss, particularly during acute flares. These findings highlight the broader socioeconomic impact of GPP beyond direct costs. Strategies to enhance patients’ presenteeism, reduce absenteeism, and prevent unemployment are likely to have the greatest impact on reducing societal costs. Background Generalized Pustular Psoriasis (GPP) constitutes a rare yet severe variant of psoriasis, characterised by lakes of sterile pustules on erythematous skin, with sudden flares, and often accompanied by potentially life-threatening systemic symptoms (Choon et al., 2023). GPP is often associated with activation of the interleukin-36 (IL-36) pathway and excessive neutrophil infiltration (Hawkes et al., 2023; Sugiura, 2022). The presence of non-cutaneous symptoms such as fever, malaise, and joint pain highlights the systemic nature of GPP (Reynolds et al., 2022). The unpredictable nature of GPP exacerbations contributes to heightened psychological distress beyond the physical discomfort of the lesions (Hoegler et al., 2018; Reisner et al., 2022). According to the 12th Report of the Malaysian Psoriasis Registry (MPR), GPP accounts for about 1.6% of patients with psoriasis, and over 60% of those with pustular psoriasis (Robinson et al., 2020). Data on the prevalence and incidence rate of GPP worldwide remains scarce and varies across geographical regions. In Japan, the estimated prevalence stands at 7.46 per million and as low as 1.76 per million in France (Augey et al., 2006; Ohkawara et al., 1996). In contrast, the prevalence in Korea was reported as high as 122.3 per million (Lee et al., 2017). Although rare, GPP has a significant impact on affected individuals, greatly reducing their quality of life. During the flare stage, patients experienced poorer quality of life, as indicated by a significantly higher mean DLQI and a lower EQ-5D score compared with those without flares (Seah et al., 2025). Additionally, several studies have reported a significantly higher economic burden among GPP patients than among patients with plaque psoriasis and the general population. The cost-driving components were primarily higher medication costs, more frequent inpatient admissions, and outpatient visits attributable to its poor prognosis, complications and comorbidities (Choon et al., 2022a; Hanna et al., 2021; Morita et al., 2021). However, while these studies emphasise the direct medical costs, evidence on indirect costs, such as productivity loss, absenteeism, and diminished work capacity, remains limited in Malaysia. Addressing this gap is critical, as indirect costs can represent a significant share of the overall disease burden, underscoring the novelty and importance of the present study. However, until now, there are limited data on the indirect cost in monetary terms related to GPP in Malaysia, even globally. A study in Sweden found that sick leave among individuals with GPP was already higher than that of the general population even before diagnosis (Löfvendahl et al., 2023). These findings highlight the importance of understanding indirect costs to assess the economic burden of GPP comprehensively. Indirect costs are often associated with unemployment, absenteeism, presenteeism, and patients’ out-of-pocket expenses, which are often neglected in the Malaysian healthcare system. A comprehensive evaluation across this spectrum is essential to accurately estimate the economic burden of GPP in the country. The study addressed the research gap by measuring work productivity loss, including absenteeism, presenteeism, and overall work impairment, as well as patients’ out-of-pocket expenses that are not currently captured from the perspective of the Malaysian Ministry of Health (MOH), such as transportation and supplement expenses. It also aimed to identify factors associated with the economic burden of GPP to guide the design and implementation of targeted interventions and policies to minimise the burden of GPP in Malaysia. Methodology Study design The study was a primary, multicentre, cross-sectional cost analysis study conducted among patients diagnosed with GPP attending public hospitals under the Ministry of Health (MOH) in Malaysia. Study Setting This study was conducted within the Malaysian public healthcare system, a setting primarily funded through general taxation where the MOH provides heavily subsidised services. In this context, where the state bears the vast majority of direct medical costs, evaluating the indirect economic burden becomes essential to understanding the full societal and patient-level impact of the disease. The focus on adult patients within the MOH system is justified because public hospitals manage the vast majority of GPP cases in Malaysia. This is supported by national dermatology registry data, which indicates that GPP patients are predominantly treated within the public sector. Within these facilities, conventional systemic therapies (e.g. methotrexate, acitretin, and cyclosporin) are widely available, while newer targeted biologics (e.g. secukinumab, ixekizumab, guselkumab, and risankizumab) are accessible but typically reserved for patients with more severe disease or inadequate response to conventional treatments. To ensure a nationally representative cohort, six tertiary hospitals were selected to mirror Malaysia’s centralised referral structure. These study sites, including Hospital Pulau Pinang (HPP) and Hospital Raja Permaisuri Bainun (HRPB) in the North; Hospital Sultanah Aminah (HSA) and Hospital Sultan Ismail (HSI) in the South; Hospital Kuala Lumpur (HKL) in the Central region; and Hospital Queen Elizabeth (HQE) in East Malaysia, serve as the dermatology referral centres for their respective regions. Multicentre selection in this study ensures a geographically and demographically diverse patient population that reflects the national clinical landscape for GPP management. Refer to Figure 1 for the dermatology referral centres in Malaysia. Figure 1.Geographical distribution of dermatology referral centres in Malaysia.A map of Malaysia marking 6 dermatology referral hospitals at Penang, Perak, Kuala Lumpur, Johor, and Kota Kinabalu with labels.The figure shows a labeled map of Malaysia highlighting dermatology referral centres involved in the study. Geographical distribution of dermatology referral centres in Malaysia. Study participant Patients were eligible for inclusion if they (i) were aged 18 years or older, (ii) had a confirmed diagnosis of GPP by a qualified dermatologist based on ERASPEN criteria, and (iii) were attending dermatology clinics at MOH public hospitals in Malaysia during the recruitment period (15 May 2024 to 30 November 2024). A purposive and convenience sampling strategy was employed to ensure a diverse representation of disease profiles. Patients were excluded if they (i) had defaulted on their regular follow-up appointments within the past year, (ii) did not have a forthcoming scheduled dermatology clinic appointment, or (iii) were unable to participate in the interview session. The inclusion criteria for this study were designed to ensure both methodological rigour and economic relevance. The age threshold of 18 years and older was specifically established to focus on the working-age population, as this demographic is the most directly impacted by indirect costs such as absenteeism, presenteeism, and long-term employment instability. Furthermore, the requirement for a confirmed diagnosis via the ERASPEN criteria was essential to maintain diagnostic consistency across the study sites, minimising misclassification and ensuring that the data reflected true GPP cases. The exclusion criteria were established to ensure data integrity and reliability. Patients who had defaulted on regular follow-up within the past year were excluded to ensure the availability of complete clinical records and to minimise recall bias regarding historical flare frequency and healthcare resource use. As the study relied on direct patient interviews to assess productivity using the WPAI tool, a scheduled clinic appointment and the ability to participate in the interview were required for practical data collection. While this convenience sampling approach was chosen for feasibility, the multicentre design was utilised to mitigate localised selection bias. Study instrument The human capital approach (HCA) was used to quantify the value of productivity loss, comprising unemployment or early retirement, changes from permanent to part-time jobs, and overall work impairment, derived from the WPAI tool for the employed population. According to the Malaysia Department of Statistics, the average monthly wage reported was MYR 2,844.00 per month, the daily wage was MYR 142.20, and the hourly wage was MYR 17.80, assuming 5 working days per week and 160 working hours per month (Department of Statistics, 2025, n.d.). Work productivity loss was estimated by examining three key components. First, the analysis considered the impact of unemployment or early retirement due to GPP. Second, it evaluated the transition from permanent to part-time employment, which can signal a reduction in work capacity. Third, the overall work impairment derived from the Work Productivity and Activity Impairment (WPAI) tool was used for the employed population, which included absenteeism and presenteeism. The WPAI tool is a validated instrument to measure impairments over the past 7 days in both paid work and unpaid work due to one’s health. The WPAI instrument was used in this study in accordance with the developers’ policy, which specifies that no permission or license is required for non-commercial research use. It comprises six questions (Q1-Q6), three related to absenteeism (work time loss due to one’s health), one related to presenteeism (reduced productivity during work due to one’s health) and two items related to activity impairments (Reilly et al., 1993). Refer to Table 1 for descriptions and scoring for WPAI questionnaire. Table 1.WPAI questions and scoring.QuestionFocus AreaItem DescriptionScoring/CalculationQ1Employment statusAre you currently employed?Determines applicability of Q2–Q5Q2Absenteeism (health-related)Hours missed from work due to health in past 7 daysUsed to calculate absenteeism %Q3Absenteeism (non-health reasons)Hours missed from work for non-health reasons (e.g. vacation, holidays)Excluded from absenteeism calculationQ4Actual hours workedNumber of hours actually worked in past 7 daysCombined with Q2 to calculate total available work hoursQ5PresenteeismHow much did health problems affect productivity while at work? (0 = no effect, 10 = completely prevented work)Presenteeism % = Q5 ÷ 10Q6Activity impairmentHow much did health problems affect daily activities outside work? (0 = no effect, 10 = completely prevented activities)Activity impairment % = Q6 ÷ 10 Table 1.WPAI questions and scoring.QuestionFocus AreaItem DescriptionScoring/CalculationQ1Employment statusAre you currently employed?Determines applicability of Q2–Q5Q2Absenteeism (health-related)Hours missed from work due to health in past 7 daysUsed to calculate absenteeism %Q3Absenteeism (non-health reasons)Hours missed from work for non-health reasons (e.g. vacation, holidays)Excluded from absenteeism calculationQ4Actual hours workedNumber of hours actually worked in past 7 daysCombined with Q2 to calculate total available work hoursQ5PresenteeismHow much did health problems affect productivity while at work? (0 = no effect, 10 = completely prevented work)Presenteeism % = Q5 ÷ 10Q6Activity impairmentHow much did health problems affect daily activities outside work? (0 = no effect, 10 = completely prevented activities)Activity impairment % = Q6 ÷ 10 WPAI questions and scoring. The calculation for annual work impairment cost: Annual Work Impairment Cost=Overall Work Impairment %×Monthly Wage×12 Out-of-pocket expenses comprised two main components. The first involved the transportation and travel expenses incurred during clinic follow-ups, while the second related to the costs associated with self-purchasing supplements or medications. All currency conversions in this study were conducted using exchange rates published by Bank Negara Malaysia on 4 October 2025. Specifically, the rates applied were MYR 1 = USD 0.2373 (USD 1 = MYR 4.21) and MYR 1 = EUR 0.1765 (EUR 1 = MYR 5.67) (Bnm.Gov., n.d.). These rates were selected to ensure consistency and reliability when converting costs reported in Malaysian Ringgit (MYR) into United States Dollars (USD) and Euros (EUR), thereby facilitating international comparison and interpretation of the findings. Study sample size The sample size was calculated using the following formula: Sample size=[z2∗p∗(1−p)]/d2 where z = 1.96 (corresponding to a 95% confidence interval) d = 0.28 (the margin of error) p = 198 per million individuals (prevalence of GPP between year 2010 and 2020 in Malaysia derived from published study by Choon et al. (2023)) Based on this formula, the calculated minimum sample size is 8. The sample size was derived using prevalence-based formulas for rare diseases, which indicated feasibility within the constraints of the available patient population. However, given the rarity of GPP, this sample remains relatively small and should be interpreted with caution. Statistical analysis Data entry was performed in Microsoft Excel, and statistical analyses were conducted using IBM SPSS Statistics version 28.0 (IBM, Armonk, NY, USA). Based on the distribution of data, the continuous variables were reported as means [standard deviation (SD)] or medians [interquartile range (IQR)]. The categorical variables were presented as numbers and percentages. Additionally, cost estimates were presented as mean (SD) and median (IQR) values where appropriate. A summary descriptive analysis was performed on demographic variables (age, gender, state, marital status, ethnicity, BMI, and educational level). Univariate analysis using the Mann–Whitney U test was conducted to compare WPAI-derived outcomes, including absenteeism, presenteeism, overall work impairment, and the annual work impairment cost, between patients experiencing flares and those in non-flare periods. Univariable and multivariable linear regression analyses were conducted to determine the predictors of indirect cost estimation. The model was controlled for confounding variables such as gender, age, ethnicity, underlying comorbidities. Statistical significance was set at the value of p < 0.05 at a 95% confidence interval. Ethical approval The study was registered with the National Medical Research Register and obtained ethical approval from the Medical Research and Ethics Committee (MREC) (NMRR ID-24-00798-OQ3). Informed consent was obtained from all participants before being included in the study. Participants were provided with comprehensive information about the study's objectives, their role in the research, and their right to withdraw without any consequences. Written consent was obtained for both participation and the use of anonymised data in any resulting publications. Results A total of 54 patients with GPP were enrolled in the study, with the majority being female (n = 41, 75.93%). The mean age of the patients was 44.56 years (SD: 15.45 range: 18-74). Majority of the patients were recruited from Johor (n = 26, 48.15%), followed by Sabah (n = 8, 14.81%) and Pulau Pinang (n = 8, 14.81%). The mean BMI was 27.14 (SD: 6.42), and 33.33% (n = 18) of the patients was classified as obese. The average duration of the disease was 11.18 years (SD: 9.81). The detailed patients’ characteristics are illustrated in Table 2. Table 2.Patient sociodemographic and clinical characteristics (n = 54).Parametern (%)Age (years), mean (SD)44.56 (15.45)GenderMaleFemale13 (24.07)41 (75.93)StateJohorPulau PinangSabahPerakSelangorKedahPahang26 (48.15)8 (14.81)8 (14.81)6 (11.11)3 (5.56)2 (3.70)1 (1.85)Marital statusMarriedSingleDivorced36 (66.67)15 (27.78)3 (5.56)EthnicityMalayChineseIndianOthers26 (48.15)19 (35.19)3 (5.56)6 (11.11)BMI, (kg/m2), mean (SD)≤18.518.5-29.5≥29.527.14 (6.42)2 (3.70)34 (62.96)18 (33.33)EducationNo Formal EducationPrimary SchoolSecondary SchoolDiplomaBachelor DegreePostgraduate Degree2 (3.70)10 (18.52)24 (44.44)9 (16.67)8 (14.81)1 (1.85)Employment statusEmployedNon-employedRetiredStudent29 (53.70)22 (40.74)1 (1.85)2 (3.70)ComorbiditiesPlaque PsoriasisDiabetes MellitusHypertensionDyslipidaemiaPsoriatic ArthritisBronchial Asthma30 (55.56)9 (16.67)15 (27.78)12 (22.22)14 (25.93)3 (5.56)Duration of confirmed GPP diagnosis,Years, mean (SD)11.18 (9.81)Treatment of GPPTopicalOral SystemicInjection36 (66.67)32 (59.26)20 (37.04)Flare statusFlareNon-flare13 (24.07)41 (75.93) Table 2.Patient sociodemographic and clinical characteristics (n = 54).Parametern (%)Age (years), mean (SD)44.56 (15.45)GenderMaleFemale13 (24.07)41 (75.93)StateJohorPulau PinangSabahPerakSelangorKedahPahang26 (48.15)8 (14.81)8 (14.81)6 (11.11)3 (5.56)2 (3.70)1 (1.85)Marital statusMarriedSingleDivorced36 (66.67)15 (27.78)3 (5.56)EthnicityMalayChineseIndianOthers26 (48.15)19 (35.19)3 (5.56)6 (11.11)BMI, (kg/m2), mean (SD)≤18.518.5-29.5≥29.527.14 (6.42)2 (3.70)34 (62.96)18 (33.33)EducationNo Formal EducationPrimary SchoolSecondary SchoolDiplomaBachelor DegreePostgraduate Degree2 (3.70)10 (18.52)24 (44.44)9 (16.67)8 (14.81)1 (1.85)Employment statusEmployedNon-employedRetiredStudent29 (53.70)22 (40.74)1 (1.85)2 (3.70)ComorbiditiesPlaque PsoriasisDiabetes MellitusHypertensionDyslipidaemiaPsoriatic ArthritisBronchial Asthma30 (55.56)9 (16.67)15 (27.78)12 (22.22)14 (25.93)3 (5.56)Duration of confirmed GPP diagnosis,Years, mean (SD)11.18 (9.81)Treatment of GPPTopicalOral SystemicInjection36 (66.67)32 (59.26)20 (37.04)Flare statusFlareNon-flare13 (24.07)41 (75.93) Patient sociodemographic and clinical characteristics (n = 54). Total mean annual indirect costs The total mean annual indirect costs per patient with GPP was estimated to be MYR 12,502.02 (SD: MYR 13,273.83), equivalent to approximately USD 2,967 and EUR 2,207 (based on exchange rates of MYR 1 = USD 0.2373 and MYR 1 = EUR 0.1765) comprising work productivity loss (unemployment or early retirement due to GPP, transitioning from permanent to part-time employment due to GPP, and overall work impairment derived from the WPAI tool) and out-of-pocket expenses (transportation and travel expenses to clinic follow-up and self-purchase supplements or medications). The total annual indirect costs per patient were calculated as the sum of total indirect costs for each component in work productivity loss and out-of-pocket expenses, divided by the total population enrolled in the study (n = 54). Further details are illustrated in Table 3. Table 3.Total annual indirect cost associated with GPP.Sources of Costn* (%)Average /patient/year (MYR)Average Annual Productivity Loss (duration in months/hours/percentage)Annual Indirect Cost(MYR)** (n = 54)Work Productivity LossUnemployed group:Unemployment/early retirement due to GPPJob changed (permanent to parttime) due to GPPEmployed group: WPAI outcomeAbsenteeism aPresenteeism bOverall work impairment cOut-of-Pocket ExpensesTransportation and travel expensesSelf-purchase supplements/medications8 (14.81)3 (5.56)29 (53.70)53 (98.15)27 (50.00)16,459.6520,826.005,342.8011,415.2312,762.54888.642396.835.79 months1,170 h15.66%33.45%37.40%--131,677.2062,478.00370,113.7047,097.8064,714.30Total annual indirect costTotal annual indirect cost per patient676,081.0012,520.02*n denotes the number of patients who reported the sources of cost described in the table.**Average annual cost was derived by translating hours of loss productivity into MYR, with a median monthly income of MYR 2,844 and hourly income of MYR 17.80.aAbsenteeism (A) = Q2/(Q2 + Q4); b Presenteeism (B) = Q5/10; c Overall work impairment = A + [(1-A) x B], derived from WPAI tool (Reilly et al., 1993). Table 3.Total annual indirect cost associated with GPP.Sources of Costn* (%)Average /patient/year (MYR)Average Annual Productivity Loss (duration in months/hours/percentage)Annual Indirect Cost(MYR)** (n = 54)Work Productivity LossUnemployed group:Unemployment/early retirement due to GPPJob changed (permanent to parttime) due to GPPEmployed group: WPAI outcomeAbsenteeism aPresenteeism bOverall work impairment cOut-of-Pocket ExpensesTransportation and travel expensesSelf-purchase supplements/medications8 (14.81)3 (5.56)29 (53.70)53 (98.15)27 (50.00)16,459.6520,826.005,342.8011,415.2312,762.54888.642396.835.79 months1,170 h15.66%33.45%37.40%--131,677.2062,478.00370,113.7047,097.8064,714.30Total annual indirect costTotal annual indirect cost per patient676,081.0012,520.02*n denotes the number of patients who reported the sources of cost described in the table.**Average annual cost was derived by translating hours of loss productivity into MYR, with a median monthly income of MYR 2,844 and hourly income of MYR 17.80.aAbsenteeism (A) = Q2/(Q2 + Q4); b Presenteeism (B) = Q5/10; c Overall work impairment = A + [(1-A) x B], derived from WPAI tool (Reilly et al., 1993). Total annual indirect cost associated with GPP. *n denotes the number of patients who reported the sources of cost described in the table. **Average annual cost was derived by translating hours of loss productivity into MYR, with a median monthly income of MYR 2,844 and hourly income of MYR 17.80. Absenteeism (A) = Q2/(Q2 + Q4); b Presenteeism (B) = Q5/10; c Overall work impairment = A + [(1-A) x B], derived from WPAI tool (Reilly et al., 1993). Work productivity loss Work productivity loss was estimated using three components: unemployment or early retirement due to GPP; transitioning from permanent to part-time employment due to GPP; and overall work impairment, derived from the WPAI tool, among the employed population. A minority of patients reported unemployment or early retirement (15%), or a shift to part-time work (6%) due to GPP, indicating the disease’s impact on long-term employability. Among those employed, work productivity loss was substantial, with overall work impairment averaging 37%, mainly driven by presenteeism rather than absenteeism. Activity impairment was also high at 36%, though not converted into monetary terms, due to the lack of a standardised conversion method. Approximately half of the respondents (54%, n = 29) were employed. For these individuals, absenteeism, presenteeism, and overall work impairment were assessed using the WPAI tool. Presenteeism accounted for a greater share of productivity loss than absenteeism across all disease severity levels. On average, absenteeism was about 16%, presenteeism about 33%, and overall work impairment approximately 37%. In monetary terms, this translated into an annual indirect cost of roughly MYR 12,760 per employed patient, based on average monthly income levels. Out-of-pocket expenses Out-of-pocket expenses consisted of two components: transportation and travel expenses to clinic follow-up and self-purchase of supplements or medications. The majority of the patients, 53 out of 54 patients (98.15%), reported incurring transportation and travel expenses for clinic follow-ups due to GPP, except for one patient who reported no expenses due to the use of public sector home-care services. The estimated total annual cost per patient for transportation and travel was MYR 888.64, including public transport, petrol, parking, toll or hired taxi services. Additionally, 50% (n = 27) of the patients reported self-purchasing supplements or medications, primarily vitamin C and moisturiser to maintain skin hydration, which were not recommended by physicians but initiated on their own. The average cost estimation per patient for these self-purchased items was MYR 2,396.83, equivalent to approximately USD 569 and EUR 423. Table 3 provides a summary of the total annual indirect cost comprised of work productivity loss and out-of-pocket expenses. Cost during GPP flares The study compared work productivity and activity impairment between flare and non-flare periods, revealing that both work and activity impairment were significantly higher during GPP flares compared to non-flare periods. During GPP flares, productivity losses were substantial: absenteeism averaged around 40%, while presenteeism and overall work impairment rose to approximately 83% and 84%, respectively. In comparison, during non-flare periods, absenteeism dropped to about 8%, with presenteeism at 18% and overall work impairment at 22%. Activity impairment followed a similar pattern, increasing to 78% during flares compared with 23% during non-flare periods. During GPP flares, work and activity productivity were markedly reduced compared with non-flare periods, with absenteeism, presenteeism, activity impairment, and overall work impairment all showing significant increases (p < 0.05 for absenteeism; p < 0.001 for the others). In monetary terms, the annual work impairment cost per patient was almost four times higher during flare periods (approximately MYR 28,800) than during non-flare periods (about MYR 7,700), representing an average difference of MYR 23,900 (95% CI: MYR 14,700–28,800; p < 0.001, Mann–Whitney U test). Further details are illustrated in Table 4, Figures 2 and 3 among patients with GPP flares and non-flare- at the interview time. Figure 2.Work Productivity and Activity Impairment derived WPAI.A bar chart comparing Work Productivity and Activity Impairment in GPP flares versus non flares, with higher values for flares throughout.The figure shows a vertical bar chart comparing Work Productivity and Activity Impairment in patients with generalized pustular psoriasis flares and non flares at the interview time. The x axis lists 4 outcomes from left to right: Absenteeism, Presenteeism, Overall Work Impairment, and Activity Impairment. The y axis is labeled Work Productivity and Activity Impairment in percent, ranging from 0.00 percent to 90.00 percent in 10.00 percent intervals. For each outcome, the chart displays 2 bars: patients in non flares and patients in flares. For Absenteeism, non flares are 7.76 percent and flares are 40.47 percent. For Presenteeism, non flares are 17.73 percent and flares are 82.86 percent. For Overall Work Impairment, non flares are 22.47 percent and flares are 84.31 percent. For Activity Impairment, non flares are 23.17 percent and flares are 77.69 percent. Upward arrows above each pair show approximate differences of 32.71 percent, 65.13 percent, 61.84 percent, and 54.52 percent respectively. All data are approximate. Figure 3.Annual work impairment cost per patient according to GPP flare status.A bar chart comparing annual work impairment cost per patient between non generalized pustular psoriasis flares and flares.The figure shows a vertical bar chart summarizing annual work impairment cost per patient. The horizontal axis lists 2 groups: During non generalized pustular psoriasis flares with sample size 22, and During generalized pustular psoriasis flares with sample size 7. The vertical axis is labeled Annual Work Impairment Cost per Patient in Malaysian ringgit. The numeric scale on the vertical axis runs from 0.00 to 35,000.00 in increments of 5,000.00. The bar for the non flare group reaches approximately 7,668.60 Malaysian ringgit. The bar for the flare group reaches approximately 28,772.06 Malaysian ringgit. A vertical arrow between the tops of the 2 bars is annotated with the value 3.75, indicating the approximate fold difference between the 2 annual cost values. The flare group bar is visibly taller than the non flare group bar, indicating higher annual work impairment costs during generalized pustular psoriasis flares compared with non flare periods. All data are approximate. Table 4.Cost comparison derived from WPAI according to GPP flare status for the employed cohort.WPAI outcomeIndirect cost for employed cohort (percentage/MYR)During non-GPP flare(n = 22)During GPP flare(n = 7)Total employed cohort(n = 29)P-valuesAbsenteeismaPresenteeismbOverall work impairmentc7.76%17.73%22.47%40.47%82.86%84.31%15.66%33.45%37.40%0.048*<0.001*<0.001*Annual work impairment cost168,709.25201,404.45370,113.70<0.001*Annual work impairment cost per patient7,668.6028,772.0612,762.54<0.001*aAbsenteeism (A) = Q2/(Q2 + Q4).bPresenteeism (B) = Q5/10.cOverall work impairment = A + [(1-A) x B], derived from WPAI tool (Reilly et al., 1993).*Statistical significance at p ≤ 0.05 using Mann-Whitney U test. Work Productivity and Activity Impairment derived WPAI. Annual work impairment cost per patient according to GPP flare status. Table 4.Cost comparison derived from WPAI according to GPP flare status for the employed cohort.WPAI outcomeIndirect cost for employed cohort (percentage/MYR)During non-GPP flare(n = 22)During GPP flare(n = 7)Total employed cohort(n = 29)P-valuesAbsenteeismaPresenteeismbOverall work impairmentc7.76%17.73%22.47%40.47%82.86%84.31%15.66%33.45%37.40%0.048*<0.001*<0.001*Annual work impairment cost168,709.25201,404.45370,113.70<0.001*Annual work impairment cost per patient7,668.6028,772.0612,762.54<0.001*aAbsenteeism (A) = Q2/(Q2 + Q4).bPresenteeism (B) = Q5/10.cOverall work impairment = A + [(1-A) x B], derived from WPAI tool (Reilly et al., 1993).*Statistical significance at p ≤ 0.05 using Mann-Whitney U test. Cost comparison derived from WPAI according to GPP flare status for the employed cohort. Absenteeism (A) = Q2/(Q2 + Q4). Presenteeism (B) = Q5/10. Overall work impairment = A + [(1-A) x B], derived from WPAI tool (Reilly et al., 1993). *Statistical significance at p ≤ 0.05 using Mann-Whitney U test. Absenteeism, presenteeism, and overall work impairment are relevant only to the employed population, whereas activity impairment applies to the entire population. Predictors of cost A simple linear regression was conducted to examine the predictors of total mean annual indirect cost, including age, state, gender, marital status, ethnicity, BMI, educational level, employment status, comorbidities, duration of confirmed GPP diagnosis, medication type and GPP flare status. Employment status was significantly associated with total annual indirect costs (p < 0.01), with a beta coefficient of −15,544.13 (95% CI: −22,603.61, −8,484.66). The presence of dyslipidaemia as a comorbidity was also significantly associated with total annual indirect costs (p = 0.046), with a beta coefficient of −9,752.76 (95% CI: −19,304.60, −200.92). Besides, GPP flare status was significantly associated with total annual indirect costs, with a beta coefficient of −14,200.66 (95% CI: −23,011.03, −5,390.29). Further details are illustrated in Table 5. Table 5.Univariable linear regression for predictors of total annual indirect cost.VariablesF-StatisticUnstandardised Coefficient (B)95% Confidence Intervalp-valueLower boundUpper boundAge3.044−227.97−490.1534.210.087Gender1.004−4,775.94−14,340.034,788.150.321State0.025240.71−2,802.353,283.440.874Marital status0.144−1,651.05−10,396.347,091.250.706Ethnicity0.0002,121.95−4,248.004,268.000.996BMI2.335483.68−151.491,018.860.133Educational level0.151−658.40−4,056.222,739.410.699Employment status19.52−15,544.13−22,603.61−8,484.66<0.001*ComorbiditiesPlaque psoriasisDiabetes mellitusHypertensionDyslipidaemiaPsoriatic arthritisBronchial asthma1.1182.072.4974.1980.1570.0004,330.46−7,788.76−7,099.88−9,752.76−1,857.60−8.24−3889.63−18,561.92−16,093.10−19,304.60−11,263.73−18,030.7212,550.563,074.401,913.33−200.927,548.5518,014.240.2950.1560.1200.0460.6940.999Duration of confirmed GPP diagnosis0.02231.69−392.98456.360.882Medication typeTopicalOral systemicInjection0.6932.8173.2433,610.084,077.98−7,443.34−5,089.46−1,338.0215,737.4312,309.6215,028.11850.760.4090.0990.078GPP flare status10.461−14,200.66−23,011.03−5,390.290.002*Boldness indicated that the variable is significant, p-value<0.05 with 95% confidence interval. Table 5.Univariable linear regression for predictors of total annual indirect cost.VariablesF-StatisticUnstandardised Coefficient (B)95% Confidence Intervalp-valueLower boundUpper boundAge3.044−227.97−490.1534.210.087Gender1.004−4,775.94−14,340.034,788.150.321State0.025240.71−2,802.353,283.440.874Marital status0.144−1,651.05−10,396.347,091.250.706Ethnicity0.0002,121.95−4,248.004,268.000.996BMI2.335483.68−151.491,018.860.133Educational level0.151−658.40−4,056.222,739.410.699Employment status19.52−15,544.13−22,603.61−8,484.66<0.001*ComorbiditiesPlaque psoriasisDiabetes mellitusHypertensionDyslipidaemiaPsoriatic arthritisBronchial asthma1.1182.072.4974.1980.1570.0004,330.46−7,788.76−7,099.88−9,752.76−1,857.60−8.24−3889.63−18,561.92−16,093.10−19,304.60−11,263.73−18,030.7212,550.563,074.401,913.33−200.927,548.5518,014.240.2950.1560.1200.0460.6940.999Duration of confirmed GPP diagnosis0.02231.69−392.98456.360.882Medication typeTopicalOral systemicInjection0.6932.8173.2433,610.084,077.98−7,443.34−5,089.46−1,338.0215,737.4312,309.6215,028.11850.760.4090.0990.078GPP flare status10.461−14,200.66−23,011.03−5,390.290.002*Boldness indicated that the variable is significant, p-value<0.05 with 95% confidence interval. Univariable linear regression for predictors of total annual indirect cost. Boldness indicated that the variable is significant, p-value<0.05 with 95% confidence interval. Additionally, a multiple linear regression analysis was conducted for those variables significantly associated with the total annual indirect costs in the simple linear regression analysis (i.e. patients’ employment status, GPP flare status and presence of dyslipidaemia) to further examine the relationship between total annual indirect cost and variables. Patients in their GPP flares and being employed were both found to be associated with higher indirect costs (p < 0.001). The overall model was statistically significant, F (3, 50) = 13.45, p < .001, and accounted for approximately 41% of the variance in indirect cost (adjusted R² = .409). GPP flare and employment status were statistically significant predictors of higher total indirect cost, with patients experiencing flares and those who are employed incurring higher costs. Further details are illustrated in Table 6. Table 6.Multivariable analysis of predictors of the total indirect cost estimation.VariablesUnstandardised Coefficient (B)Standardised Coefficient (Beta)P-value95% Confidence IntervalAdjusted R2 = 0.409Lower BoundUpper BoundDyslipidemia−3,084.90−0.0860.436−10,973.444,803.65Employment status−14,963.73−0.503<0.001−21,412.38−8,515.12Flare Status−13,582.23−0.391<0.001−21,068.05−6,096.41Boldness indicated that the variable is significant, p-value<0.05 with 95% confidence interval. Table 6.Multivariable analysis of predictors of the total indirect cost estimation.VariablesUnstandardised Coefficient (B)Standardised Coefficient (Beta)P-value95% Confidence IntervalAdjusted R2 = 0.409Lower BoundUpper BoundDyslipidemia−3,084.90−0.0860.436−10,973.444,803.65Employment status−14,963.73−0.503<0.001−21,412.38−8,515.12Flare Status−13,582.23−0.391<0.001−21,068.05−6,096.41Boldness indicated that the variable is significant, p-value<0.05 with 95% confidence interval. Multivariable analysis of predictors of the total indirect cost estimation. Boldness indicated that the variable is significant, p-value<0.05 with 95% confidence interval. Discussion This study is the first in Malaysia to estimate the indirect cost burden of GPP encompassing six study sites representing the various regions of the Malaysian population. These selected study sites reflect the referral structure of Malaysia’s health system and capture diverse patient populations. Direct comparisons with other countries remain difficult. Few studies have quantified the indirect costs of GPP specifically, reflecting its rarity. Most economic data derive from plaque psoriasis or psoriatic arthritis, which are more prevalent and better studied. Nevertheless, research from the US and Sweden has shown that productivity losses represent a considerable proportion of the overall economic burden in psoriasis (Vanderpuye-Orgle et al., 2015) (Svedbom et al., 2016). In our study, after considering variables including patients’ demographics (age, gender, marital status, state, ethnicity, BMI, educational level) and patients’ clinical characteristics (flare status, comorbidities, duration of confirmed GPP diagnosis), we found that patients in their GPP flares and employment were associated with higher total annual indirect cost. Besides, among employed patients, productivity loss was driven more by presenteeism than absenteeism, highlighting the substantial impact of GPP on individuals who remain in the workforce. This pattern mirrors findings from multinational studies of psoriasis, which consistently report higher levels of presenteeism across European countries and the United States (Villacorta et al., 2020). These findings suggest that the economic burden of GPP extends beyond lost workdays to reduced effectiveness while at work, an aspect often under-recognised in economic evaluations (Villacorta et al., 2020; Lebwohl et al., 2022). In view of GPP being a rare disease with an unpredictable nature of flares (Choon et al., 2022), further analysis of work productivity and activity impairment was undertaken using the WPAI tool. The analysis showed that both work and activity impairment were considerably greater during flare periods than during remission. These results align with recent regional observations in Southeast Asia, where GPP often leads to prolonged flare durations and frequent hospitalisations (Jullien et al., 2023; Choon et al., 2013; Svedbom et al., 2016). In the Malaysian context, employment instability is a critical driver, as evidenced by unemployment rates near 15% among the GPP cohort in this study. The nature of this burden is distinct across sectors; for patients in the formal employment, frequent medical leave necessitated by unpredictable flares may result in reduced career progression and diminished long-term earning potential. Conversely, those in informal employment face immediate and total income loss during flare-induced absences, creating profound financial vulnerability. While conventional therapies remain highly subsidised and accessible, restricted access to biologics and newer targeted therapies may lead to suboptimal disease control (Seah, et al., 2025). This necessitates frequent treatment switching and contributes to more frequent acute hospital admissions, both of which exacerbate overall productivity loss and deepen the economic impact on patients and the country. The integration of newer, targeted therapies specifically designed for GPP, such as interleukin-36 (IL-36) receptor inhibitors, represents a pivotal shift in the therapeutic landscape with significant implications for health policy. Although public sector reimbursement frameworks currently favour conventional systemic therapies, prioritisation of targeted therapies may significantly mitigate the substantial indirect cost burden through rapid flare resolution and the prevention of acute hospitalisations. The study methodology shares common limitations with typical survey-based studies, including the possibility of recall bias and selection bias. Besides, all the data were estimated only at one time point, which could lead to potential bias limitations. In order to improve the strength of the study, efforts were made to design a comprehensive survey that captured detailed and specific data to minimise recall bias. Furthermore, the study employed a standardised methodology for data collection and analysis to enhance consistency and reliability. The estimates should also be interpreted in light of methodological considerations. Different approaches yield varying results: the human capital approach assumes all lost work time represents societal burden, whereas the friction cost approach limits losses to the period before a replacement is found (Kawalec & Malinowski, 2015; Krol & Brouwer, 2014). These methodological constraints, combined with the scarcity of GPP-specific economic data, limit the comparability and generalisability of our indirect cost estimates. Besides, the sample size was relatively small (n = 54), so the findings cannot be considered fully generalisable. Larger, multi-centre studies, ideally leveraging national registries or international collaborations, are needed to validate and extend these results. Such efforts would provide more robust evidence on the economic burden of GPP and strengthen the basis for health-system planning and policy development. Furthermore, a key limitation of this study is that only the dosage form (e.g. oral or injectable) was captured, and detailed information on individual medications, dosing frequency, and treatment duration was unavailable. This limitation should be addressed in future research to provide a more comprehensive understanding of treatment utilisation. Despite the above limitations, this study is the first multi-centre investigation in Malaysia to quantify the indirect costs of GPP, incorporating data from diverse regions of the country. It assessed both productivity loss and out-of-pocket expenses, such as transportation and self-purchased treatments, domains that are often underrepresented in economic evaluations. By extending the scope beyond the WPAI tool, the study provides a more comprehensive picture of the societal burden of GPP and generates evidence that is directly relevant to health policy and clinical decision-making. Conclusion This study demonstrates that GPP imposes a substantial economic burden driven largely by work productivity loss, particularly during acute flares when productivity losses and unemployment risks are most pronounced. These findings highlight the broader socioeconomic impact of GPP beyond direct healthcare costs and support the need for effective disease management strategies that may reduce productivity loss and patient-incurred expenses. To ensure the robustness and durability of these findings, larger prospective studies are essential to validate the observed associations and clarify underlying mechanisms. Generating such evidence will better equip clinicians, policymakers, and employers to design and implement targeted interventions that reduce the societal and economic burden of GPP.