The humanistic and economic burden of atopic dermatitis among adults and adolescents in Saudi Arabia

Abstract

Aims: Atopic dermatitis (AD) is a chronic skin disease that creates a significant burden to patients and society. There is scarcity in local data about the burden of AD in the Kingdom of Saudi Arabia (KSA). We aimed to fill in this gap and quantify the humanistic and economic burden of AD among adults and adolescents in KSA.

Materials and methods: A literature search and local expert interviews were conducted to assess the disease burden. Prevalence values were estimated through the literature. International data about health-related quality of life lost owing to AD was adjusted to age and prevalence in KSA. Direct and indirect costs were calculated using a bottom-up approach. Resource utilization data were collected from local dermatologists through online interviews, and indirect costs were based on absenteeism and presenteeism estimates. Validation meetings were conducted with local experts to adjust the final estimates.

Results: The age-standardized health loss per patient due to AD is 0.187 quality-adjusted life-years (QALYs) annually, aggregating to 64 thousand lost QALYs in KSA. The annual average direct cost for a patient with AD was 2924 Saudi Riyal (SAR; 780 USD), totaling 373 million SAR in KSA (99.5 million USD). This value represents 0.2% of the annual health expenditure in KSA. The total productivity loss due to AD was 1.36 billion SAR (363.7 million USD). Overall, the economic burden of AD consumes up to 0.059% of the national gross domestic product.

Limitations: Local quality of life and productivity lost data were not available for KSA, so global averages were used, assuming these numbers also apply to KSA.

Conclusion: Indirect costs represent a large proportion of AD burden in KSA. The disease has a substantial effect on patient quality of life and social well-being. Alleviating the burden might result in significant savings in resources to society.

PLAIN LANGUAGE SUMMARY

Atopic dermatitis is one of the most common skin diseases. Mild cases of the disease cause inflamed and itchy skin, while severe cases may cause painful episodes of itching and cracked skin. Patients with atopic dermatitis and their families suffer lower quality of life as the severity of the disease increases. In countries with hot weather like Saudi Arabia, skin is more susceptible to become dry, so the disease is very prevalent. Therefore, the disease poses a significant quality of life burden as well as an economic burden due to the direct costs of treatment and the indirect costs that arise because patients become non-productive or absent from work or school. Our study aimed to quantify the economic and quality of life burden of atopic dermatitis in Saudi Arabia to understand it’s real burden and help decision makers quantify its impact on the patients and society. We conducted a literature search and interviewed local experts to determine estimates of costs and quality of life effects. The results of this study should help in prioritizing treatment disease areas in Saudi Arabia and other countries with similar circumstances.

Keywords:

• Atopic dermatitis
• burden of disease
• economic burden
• eczema
• humanistic burden
• quality of life
• Saudi Arabia

JEL Classification codes:

• C13
• C1
• C
• D61
• D6
• D

Introduction

Atopic dermatitis (AD) is a chronic inflammatory skin disease that can impose a significant burden on patients and their families¹.

Patients with AD may suffer from itching, pruritis, skin redness, and swelling. In more severe cases, the skin might crack or become scaly or lichenified. AD flares are episodes of severe itching and pain experienced by patients. These may worsen at night, resulting in restlessness and inadequate sleep for patients and their caregivers. AD is not a fatal disease, but it has a significant negative effect on patient quality of life because of its effect on daily activities. In the long term, AD predisposes patients to several psychological problems, such as depression for patients who are self-conscious about being itchy and inflamed most of the time^2–4.

Atopic dermatitis occurs as a result of multiple factors, including genetic, environmental, and immunologic. However, the pathophysiology of the disease is not very well understood^5,6. AD is usually correlated with asthma and allergic rhinitis because these diseases are all related to allergies, genetic defects, and immunologic responses toward allergens⁷. As such, asthma and allergic rhinitis are prevalent among patients with AD⁸.

Atopic dermatitis presents as mild to severe disease. The severe form of the disease can be very disrupting and resource consuming⁹. Severity is frequently assessed using questionnaires, completed by patients or their caregivers, usually asking about the frequency and intensity of flares; the area of skin affected; and the presence of redness, papules, scaling, edema or lichenification. Many of these questionnaires exist, but the most commonly used are SCORing Atopic Dermatitis (SCORAD), the Eczema Area and Severity Index (EASI), and the Atopic Dermatitis Severity Index (ADSI)^10–12.

Measuring the prevalence and incidence of AD is not an easy task; it is not diagnosed with a laboratory test or a similar objective method¹³. The disease usually develops in childhood. Symptoms may persist throughout one’s lifetime or they may improve as patients get older, sometimes resolving completely¹⁴. AD may develop for the first time in an adult patient, but this is relatively rare¹⁵. This is why the prevalence of AD is lower in adults compared with children, and most studies related to AD focus on children and adolescents because they represent a larger patient population than adults. AD is more prevalent in countries where the weather is less humid because a lack of humidity can lead to dry skin, which aggravates symptoms of the disease^16,17.

The burden of AD is usually underestimated; beyond the cost of creams and lotions, burden may affect patients in other ways. For example, AD usually flares at night, resulting in inadequate sleep and loss in productivity during the next day. These productivity losses are not limited to the patients but extend to their formal and informal caregivers as well. Furthermore, patients may experience depressive symptoms and a lack of self-confidence because of the disease. These all significantly affect patient quality of life^18–20.

Currently, there is no cure for AD, but several treatment options are available for all severity grades. Treatments usually involve topical emollients and hydrating creams to decrease skin dryness, as well as skin repairing creams. Topical corticosteroids are often used to reduce inflammation and itching, and other creams, such as topical calcineurin inhibitors, are commonly used to control immune response. Systemic immunosuppressants and anti-inflammatories may be used in more severe cases, but efficacy has been questioned^21,22. New monoclonal antibodies (e.g. dupilumab) and Janus kinase (JAK) inhibitors (e.g. upadacitinib and baricitinib) have been developed as second-line treatments for nonresponsive patients. Phototherapy may be used as a treatment option in some cases^22,23.

Previous studies that have assessed the burden of AD globally and in specific geographic areas found that the disease significantly affects the quality of life, productivity, sleep rhythm, mental health, and daily activities of patients^24–26.

In the Middle East, and in the Kingdom of Saudi Arabia (KSA) specifically, there are few publications assessing the burden of AD^27–29, but none has attempted to estimate the humanistic and economic burden in adults and adolescents. A study conducted in 2017 among female patients in Jeddah estimated the prevalence of eczema to be 16.6%³⁰. Another recent study estimated the prevalence of AD among Saudi young adults to be 13.1%²⁷.

This study aims to quantify the humanistic and economic burden of AD among adults and adolescents in KSA. The study results will provide the foundation for evidence-based decision making for policymakers and stakeholders involved in drug selection and budget allocation decisions.

Methods

A literature search and expert interviews were conducted to assess the burden of AD. A bottom-up approach was used to estimate the disease burden. To assess the humanistic burden, the annual quality-adjusted life-years (QALYs) lost due to AD was calculated by multiplying the average utility lost by one patient in KSA by the estimated number of patients. The same approach was used for costs; the average annual cost for treating one patient with AD was multiplied by the estimated number of patients to calculate the total economic burden in KSA. The study included data about adolescents and adults and excluded data about children younger than 10 years based on the World Health Organization’s adolescent definition³¹. Ethics approval was not required for this study.

During the study, a conservative approach was used at each step to avoid exaggerated values that might inflate the estimated burden. If reliable and accurate estimates were not available for a data point, the lower values of the available references were used to estimate the burden.

We expressed the humanistic burden in “QALYs lost” rather than other burden of disease measures like disability-adjusted life years (DALYs) and health-adjusted life expectancy (HALE)³² because the health gain of public health interventions or new medical interventions are standardly reported in QALYs by health technology assessment (HTA) agencies. So, “QALYs lost” value was estimated to create the potential for calculating the cost-effectiveness of such interventions at the population level. QALYs are among the few methods that allow for comparisons between interventions or across disease areas³³.

Estimating the prevalence

To estimate the humanistic and economic burden, information on the prevalence of AD in patients aged 10 years and older was required. Because quality of life and direct and indirect costs may vary by age and sex, we needed data stratified by those factors. We searched the literature for the required data using two search domains: atopic dermatitis and Saudi Arabia. Synonyms of the disease were also used to find relevant data (e.g. atopic eczema, eczema, prurigo Besnier). Because most prevalence data found in peer-reviewed journals and published studies were limited to a subgroup of patients (e.g. 13- to 14-year-olds), unstratified by age and sex, or restricted to a specific geographic area in KSA^27,30,34,35, data from the Global Burden of Disease study³⁶ for KSA were used to estimate the prevalence. The Global Burden of Disease study was searched through the Institute for Health Metrics and Evaluation (IHME) database³⁷ for patients with AD aged 10 years and older in KSA.

Humanistic burden

The annual humanistic burden of AD was defined as the number of QALYs lost for the whole population owing to AD over a 1-year period. To calculate the total QALY loss, the number of patients in each age group was multiplied by the utility lost per patient in the same age group. For example, if the 10- to 14-year-old age group includes 100 patients with AD, and each patient loses 0.1 QALYs on average owing to AD annually, the total humanistic burden for this age group would be as follows: 100 × 0.1 = 10 QALYs per year.

To calculate the QALY loss per patient, a literature search was conducted to find studies that provided utility broken down by age group for patients with AD. Quality-of-life values for adolescent subgroups aged 12–14 years and 15–17 years were retrieved from the study by Ezzedine et al.³⁸, and the quality-of-life values for adult patients aged 18–75 years were reported by Beikert et al. in seven age groups³⁹. The values in those studies were used as a starting point to calculate the QALY loss per age group.

Because the age distribution for reporting prevalence and utilities for patients with AD were not identical for all age brackets, the utility for patients aged 12–14 years was used to represent patients 10–14 years old, utility for patients aged 15–17 years was used to represent patients 15–19 years old, and utility for patients aged 18–24 years was used to represent patients 20–24 years old.

Each study measured quality of life using a different questionnaire and on different scales. Hence, conversion of all questionnaire results into one unit was needed, to allow for aggregation of results and comparison. Utility values range from 0 to 1; where 0 represents a person who is dead and 1 represents a person in full health. The EuroQoL 5-dimensions (EQ-5D) index questionnaire is a quality-of-life questionnaire that provides scores in this range; therefore, the quality-of-life scores identified were transformed to EQ-5D index scores.

The study that provided utility scores for the 12- to 14-year age group used the children’s version of the Dermatology Life Quality Index questionnaire, and the study providing scores for the 15- to 17-year age group used the adult version. The values were transformed to the EQ-5D index using an online transformation tool⁴⁰. The study that reported adult utility scores used the EQ-5D visual analog scale (VAS) questionnaire. Because the data were presented graphically, the first step was to digitize the values using the WebPlotDigitizer application⁴¹. Next, EQ-5D VAS scores were transformed to the EQ-5D index. We used the tool developed by Fasseeh et al. to transform EQ-5D VAS values to EQ-5D index values⁴². This tool was based on a linear regression conducted based on five studies that included EQ-5D VAS and EQ-5D index scores for patients with AD in the same subgroup^43–47. Fifteen data points from these studies were added to the regression model to create the data trend. The regression equation used was: $$\mathrm{EQ}-5\text{D VAS score }=\mathrm{ }0.0136\text{ * EQ}-5\text{D index score }-\mathrm{ }0.1534$$

We were able to transform all EQ-5D VAS scores to EQ-5D index scores using this equation. Finally, we had all the required age groups utility values reported into a unified score (EQ-5D index score).

Once the utilities for each age group were obtained, utility loss due to AD had to be calculated. To do this, the utility for each subgroup with AD was subtracted from the baseline utility of the average population in this age group. For example, if the baseline utility for an average person aged 10–14 years is 0.95 and the utility of the average AD patient in that same age group is 0.75, then the utility loss due to AD would be as follows: 0.95 − 0.75 = 0.2

Literature was searched to find studies reporting the baseline utilities for different age groups, either globally or in KSA specifically. The best available reference was a study reporting baseline utility for all age groups in 20 diverse countries⁴⁸. For each age group, the average of the 20 countries was calculated to represent the average utility per age group. This value was used as a proxy for age group utility in KSA. The final calculation for annual humanistic burden was as follows: $$(\text{number of patients in age group }1\mathrm{ }\times \text{ utility loss per patient in age group }1)\mathrm{ }+\mathrm{ }(\text{number of patients in age group }2\mathrm{ }\times \text{ utility loss per patient in age group }2)\mathrm{ }+\mathrm{ }(\text{number of patients in age group }3\mathrm{ }\times \text{ utility loss per patient in age group }3),\mathrm{ }\mathrm{etc}.$$

Economic burden

The economic burden of a disease represents the money or resources lost owing to the disease. We adopted a societal perspective in estimating the economic burden. The economic burden of AD is divided into direct costs, such as the cost of dermatologist visits, pharmaceuticals, hospitalization, and phototherapy sessions, and indirect costs, such as the cost of productivity loss due to the disease. Productivity loss is measured by the number of days that the patient is absent from work or school (absenteeism) and the number of days the patient is at work or school but is not productive (presenteeism)⁴⁹.

Direct costs

To calculate the direct costs locally, 2- to 3-h structured expert interviews were conducted with three clinicians who have practical knowledge and real-world experience in treating patients with AD in KSA to ask them about resource utilization (e.g. number of days of hospitalization, type of cream prescribed, frequency of using treatments). The interviews were conducted through online meetings with the experts, who had previously received a questionnaire in preparation for the meeting. Experts were chosen based on convenience sampling.

Interviews were conducted with two dermatologists, and the average values were used to alleviate the effect of individual preferences or bias. One interviewee was from King Khaled University Hospital, which is an affiliated teaching hospital located in the capital of KSA. The other interviewee was from King Abdullah Medical Complex in Jeddah, which is operating under the Ministry of Health. During the interviews, experts were asked to complete the questionnaire, assisted by the research team if clarification was needed. The structured questionnaires used in the experts’ interviews were developed through a simple literature search to define the key elements in treating AD. Because the literature revealed that AD is treated differently for patients with mild, moderate, or severe disease, the questionnaire was structured to provide values for each severity grade separately. The resource utilization value was then multiplied by the percentage of patients in a particular severity level (also provided by the experts through the interview), and all severity values were aggregated at the end. The data domains in the questionnaire for the structured interview are shown in Appendix 1 and the questionnaire template is shown in Appendix 2. For example, if 60% of the cases are mild, 30% are moderate, and 10% are severe, and if patients with a mild case use 1 unit per month, with a moderate case use 2 units per month, and with a severe case use 3 units per month, then the calculation for the final average number of units per patient per month will be as follows: $$(60\mathrm{\%}\mathrm{ }\times \mathrm{ }1)\mathrm{ }+\mathrm{ }(30\mathrm{\%}\mathrm{ }\times \mathrm{ }2)\mathrm{ }+\mathrm{ }(10\mathrm{\%}\mathrm{ }\times \mathrm{ }3)\mathrm{ }=\mathrm{ }1.5\text{ units per month}$$

Experts were also asked to provide an estimate for the unit cost of each intervention or treatment. However, this was used only as a guide for validation because the unit costs were abstracted from public prices of products in KSA. Prices were extracted either from the Saudi Food & Drug Authority official website⁵⁰, online pharmacy websites⁵¹, online shopping websites (for cosmetic creams) ⁵², or scientific publications⁵³. Data reported for patients in the older age groups were inflated using an online inflation tool for Saudi Riyal⁵⁴. For doses reported as dose per kilogram body weight, the average weight in KSA was used⁵⁵.

To calculate the total direct cost per population, the average cost per patient (from the interviews) was multiplied by the prevalence of AD in KSA. The value was adjusted to the percentage of patients diagnosed with AD⁵⁶ and did not include undiagnosed patients because patients with AD who have not been diagnosed or treated have no associated direct costs. The final total direct cost equation was: $$\text{Total direct cost }=\text{ average treatment cost per patient× number of patients in the target population× percentage of patients diagnosed}$$

Indirect costs

To calculate the effect of productivity loss on indirect costs due to AD, absenteeism and presenteeism values were calculated. The first step was to calculate the number of days lost by an average patient due to the disease. We searched the literature for relevant studies, and we identified 17 studies that reported values for the absenteeism or presenteeism of patients or their caregivers due to AD (Appendix 3). Of those studies, 16 included data about absenteeism, 9 included data about presenteeism, 15 included data about the patient burden, and only 2 included data about caregiver burden. The average annual rates of absenteeism and presenteeism for patients were calculated based on all values in the 17 studies. Because this study is a conservative one, caregiver burden was excluded from the calculation owing to scarcity of data and because caregivers were usually associated with children, and our study focused on adults and adolescents. The result of this literature search was the average annual number of days of absenteeism and presenteeism due to AD.

Next, to find the total number of days lost for the whole population due to AD, the number of productivity days lost per patient annually was multiplied by the number of employed patients. To calculate this, the prevalence of the AD population of working age (15–65 years) for male and female patients in each age group was abstracted³⁷. Then, we adjusted this value to the unemployment rate in KSA (i.e. percentage of unemployed males and females of working age), and to the labor force participation rate (LFPR; i.e. percentage of employed males and females of working age)⁵⁷.

To estimate productivity loss in the AD population in KSA, the number of days lost per patient was multiplied by both the prevalence of male and female patients of working age (adjusted to LFPR and unemployment rate) and the average daily salary. The average salary in KSA was estimated from an internet database⁵⁸. The equation we created to calculate productivity loss is as follows: $$\left[\right\{\text{LFPR male }\times \mathrm{ }(1\mathrm{ }–\text{ unemployment rate})\mathrm{ }\times \text{ prevalence male}\}\mathrm{ }+\mathrm{ }\{\text{LFPR female }\times \mathrm{ }(1\mathrm{ }–\text{ unemployment rate})\mathrm{ }\times \text{ prevalence female}\left\}\right]\mathrm{ }\times \text{ absenteeism }\times \text{ daily salary }+\mathrm{ }\left[\right\{\text{LFPR male }\times \mathrm{ }(1\mathrm{ }–\text{ unemployment rate})\mathrm{ }\times \text{ prevalence male}\}\mathrm{ }+\mathrm{ }\{\text{LFPR female }\times \mathrm{ }(1\mathrm{ }–\text{ unemployment rate})\mathrm{ }\times \text{ prevalence female}\left\}\right]\mathrm{ }\times \text{ presenteeism }\times \text{ daily salary}$$

Total economic burden

The total economic burden was calculated as the sum of direct and indirect costs.

Validation meetings

To validate the research findings, online meetings were held with four local experts in the field to ensure the findings were consistent with local settings and experiences and to make necessary adjustments based on their recommendations. Each validation meeting was managed and coordinated by two research team members (the principal researcher and a senior researcher).

Experts were chosen based on specific criteria: (1) dealing with atopic dermatitis on a daily basis, (2) representing different healthcare systems in Saudi Arabia (Ministry of Health and Population, Ministry of National Guard Health Affairs and University hospitals) and (3) representing different cities in KSA.

During these meetings, all study outcomes (utility values, direct costs, and indirect costs) and their methodology of calculations were discussed in detail with the experts. They discussed the methodology of estimation and recommended better methods or better sources for the data.

Validation meetings were recorded, and transcribed, and corrective actions were taken to adjust the estimates based on recommendations made by the experts. For example, the experts provided a more accurate local reference for the labor force population size and unemployment rate during the validation meeting, and those data were updated accordingly⁵⁵.

Results

Based on the Global Burden of Disease database in 2019 (last reported), 343,870 patients with various states of AD severity were present in KSA, representing 1.14% of the total population aged 10 years and older. Prevalence was higher among female versus male patients (1.35% vs 0.99% respectively).

Humanistic burden

On average, a patient with AD accumulates 0.69 QALYs in a life-year with the disease, which is lower than the general population. The age-standardized QALY loss per patient due to AD is 0.187 annually. Based on a total prevalence of about 344 thousand patients in KSA, and after adjusting for subgroups of patients, the total population in KSA loses about 64,000 QALYs annually owing to AD. Table 1 includes the breakdown of humanistic burden in KSA.

Table 1. Humanistic burden of atopic dermatitis in KSA.

Age group, years	Utility among patients with AD, per patient	Baseline utility for general population, per patient	Utility loss for AD, per patient	AD prevalence, no. of patients	QALYs lost annually, per population
10–14	0.76	0.93	0.17	39,970	6,724
15–19	0.70	0.93	0.23	34,702	7,841
20–24	0.77	0.93	0.15	41,643	6,446
25–34	0.73	0.92	0.18	100,224	18,438
35–44	0.71	0.90	0.19	80,067	15,414
45–54	0.68	0.86	0.18	32,629	6,006
55–64	0.54	0.82	0.28	10,496	2,953
65–74	0.71	0.80	0.09	3,154	286
≥75	0.61	0.72	0.11	985	108
Total	–	–	–	343,870	64,215

Abbreviations. AD, atopic dermatitis; KSA, Kingdom of Saudi Arabia; QALY, quality-adjusted life-year.

Direct costs

Based on the average values provided by two dermatologists via questionnaires and subsequent calculations, the estimated annual average direct cost for a patient with AD in KSA is 2,924 SAR annually. For the whole AD population in KSA, the direct costs are estimated at about 373 million SAR. This value represents approximately 0.2% of the annual health expenditure in KSA⁵⁹. Table 2 includes the details of direct costs calculations of AD in KSA.

Table 2. Direct costs of AD in KSA.

Metric	Value	Source
Annual average direct cost per patient	2,924 SAR (780 USD)	Expert questionnaires
Patients with AD aged ≥10 years, n	343,870	Global Burden of Diseases^34
Diagnosed, %	37.10	Hanifin et al.^53
Total direct cost per population 2019	373 million SAR (99.5 million USD)	Calculation
Annual healthcare expenditure	188 billion SAR (50.0 billion USD)	Global Health Observatory^56
Percentage of total healthcare expenditure in KSA	0.20	Calculation
Direct cost as a percentage of GDP	0.013	Calculation based on World Bank GDP estimate^57

Abbreviations. AD, atopic dermatitis; GDP, gross domestic product; KSA, Kingdom of Saudi Arabia; SAR, Saudi Riyal; USD, US dollars.

Indirect costs

Based on the literature search, a patient with AD is absent from work or school for an average of 6.1 days and is present but not productive for an average of 22.9 days annually owing to AD. This translates to an annual loss of up to 285 million SAR and 1079 million SAR due to absenteeism and presenteeism, respectively. The total productivity loss due to AD in KSA per year was estimated at approximately 1.36 billion SAR. This represents 0.05% of the national gross domestic product (GDP)⁶⁰. Table 3 shows the indirect costs calculation details.

Table 3. Indirect costs due to AD in KSA.

Metric	Value	Source
Average annual days of absenteeism. n	6.1	Literature search
Average annual days of presenteeism, n	22.9	Literature search
Average monthly salary	6,341 SAR (1,691 USD)	Internet database^55
Total unemployment rate, * %	7	General Authority of Statistics^54
Total LFPR, * %	61	General Authority of Statistics^54
Patients with AD aged 15–65 years, n	299,761	Global Burden of Diseases^34
Total cost due to absenteeism	285 million SAR (76 million USD)	Calculation
Total cost due to presenteeism	1,079 million SAR (288 million USD)	Calculation
Total indirect cost	1,364 million SAR (364 million USD)	Calculation
Indirect cost as a percentage of the GDP	0.046	Calculation based on World Bank GDP estimate^57

Abbreviations. AD, atopic dermatitis; GDP, gross domestic product; KSA, Kingdom of Saudi Arabia; LFPR, labor force participation rate; SAR, Saudi Riyal; USD, US dollars.

*Among patients with AD aged ≥15 years.

Total economic burden

Total economic burden (direct and indirect costs) in KSA due to AD is estimated at 1.7 billion SAR annually. Indirect costs are responsible for 79% of this burden.

Discussion

Although AD is a non-fatal disease, it carries a huge burden mainly because of the poor quality of life faced by the patients. AD is ranked 15th in disability-adjusted life-years (DALYs) lost among non-fatal diseases and first in disease burden among all skin diseases⁶¹. The age-standardized global DALYs burden of AD is comparable to other fatal diseases, such as cirrhosis and measles, and is more than double the burden for scabies and fungal skin diseases⁵⁹. The economic burden of AD in KSA might be more than seven times that of breast cancer, based on a recent retrospective cross-sectional study that estimated the annual cost burden of breast cancer at 13.3 million USD⁶² compared with 99.5 million USD for AD as estimated by our study. The methodologies of that study and our study were not exactly the same, but they are comparable in that both studies estimated the annual direct costs of the disease in KSA for adults. The cost per patient is higher for breast cancer, but the high prevalence of AD compared with breast cancer is key in assessing its higher economic burden.

When directing available resources, there may be a lower priority given to treating AD versus fatal diseases because AD mainly affects quality of life rather than years lived. This study provided an estimate of the quantitative burden of AD in Saudi Arabia to guide decision-makers in resource allocation and spending.

Concerning prevalence, the best available age-stratified data were used from the Global Burden of Disease study results; however, the values are most likely an underestimation of the real prevalence according to Saudi experts and validators. Previous studies including AD prevalence in KSA have been conducted and provided higher estimates of prevalence, but these studies were not included here either because they were not stratified by age or sex or because they were confined to a specific geographic region in KSA. The actual prevalence is likely to be higher than reported because a large portion of patients with AD are not diagnosed by a physician. Therefore, estimated values of the burden are most plausibly lower than the actual burden of AD in KSA.

The results show that approximately 64 thousand QALYs are lost annually owing to AD. This may be comparable to other more severe diseases, not only because of the physical pain and suffering that results, but also because of the significant effect that depression, inadequate sleep, and lack of self-confidence have on the quality of life of patients with AD.

The direct cost of a disease is usually what first comes to mind when burden is mentioned. AD is not a disease that usually requires surgeries, diagnostics, or expensive interventions. However, when all factors were aggregated and multiplied by the large number of patients, the burden is considerable. The fact that AD, a disease commonly perceived to be of low morbidity, consumes approximately 0.2% of the health expenditure is alarming. This sizable burden might be attributed to the sheer number of patients with AD and the high cost of medications used by patients with moderate to severe disease. This value is comparable to the contribution of the disease to the national health expenditure in Taiwan (0.314%). However, the value reported in Taiwan included both children and adults. There are substantial differences between the cost of treating mild versus moderate and severe AD; new targeted therapies can require significant expenditures because the disease is chronic.

Indirect costs are seemingly the critical issue in this disease, contributing to about 79% of the economic burden. Absenteeism and presenteeism can vary between severity levels of AD, but, on average, these are significant enough to generate an economic effect. Since we are adopting a societal perspective, both absenteeism and presenteeism contribute equally to the productivity loss due to AD. We estimated that each year, KSA will lose approximately 1.36 billion SAR as indirect costs owing to AD. This is about five times the direct cost of the disease and accounts for 0.05% of the Saudi GDP, which is one of the highest GDPs in the world⁵⁶. Notably, our study did not address the patient out-of-pocket burden related to multiple skin product trials, which are more difficult to estimate but have been reported to be significant^63,64.

The availability of local data to study disease burden is limited in the Middle East and North Africa region, due to limited accessibility to payers’ databases and patient registries⁶⁵. However, the deficiency of high-quality data should not be a barrier to research. Using the available data to conduct research should encourage local stakeholders to improve the quality of data to improve decisions eventually⁶⁵.

Limitations

Because there is scarcity of specific data regarding AD in KSA and the neighboring countries, global estimates were used, with an assumption that KSA will follow the same average. This assumption may not always be true because of cultural and climate differences. Efforts were made to adjust values to age, sex, and country, but because some values were not broken-down into these categories, the averages were used instead.

Health-related quality-of-life data were abstracted from studies that were conducted in Germany and France, and the baseline quality-of-life data came from a study that included 20 different countries. Also, we had to use estimates from more than one study because we could not identify all the quality-of-life data for the required age groups from one study. The availability of local quality-of-life data may have improved the accuracy of the results. It is important to note, then, that the quality-of-life values used here are not based on local data but on calculations and estimates.

For prevalence estimates, although the Global Burden of Disease study provided a relatively low estimate (based on Saudi experts’ opinions), it was still the best available choice because it included data broken down by age and sex. Since the estimate is probably lower than the actual value, using its values retains our study’s conservative nature.

These factors might have caused some inaccuracy in the results. However, because the intention of this study was to estimate the overall country burden of AD, it is assumed that minor inaccuracies would not cause a large deviation and would still provide a very good overview of the comparative burden of the disease. This study should be updated if better local data become available in the future.

The number of experts interviewed was small. However, to make sure the results are valid, we conducted validation meetings with local experts who confirmed the validity of the data and suggested further improvements to increase the data reliability.

Conclusions

Atopic dermatitis poses a significant burden in Saudi Arabia, especially considering the indirect cost. The total economic burden of AD consumes up to 0.059% of the Saudi GDP, which is one of the highest GDPs per country in the world⁵⁶. Also, AD has a significant effect on patient quality of life and social well-being. With optimal disease control, both humanistic and economic burdens may be reduced. Awareness about AD may also prove to be very beneficial in alleviating the huge burden. Our study provides some insight into the burden of this disease in Saudi Arabia, which has not been previously recognized.

Transparency

Author contributions

ZK, ShA, BE, ANF, MT were involved in the conception and design of the study. BE and ANF conducted the literature search. EF, BE and ANF conducted the interviews and validation meetings with the experts. MT, HD, SaA, AR, AJ and TH facilitated the interviews and the validation meetings. BE, ANF and ZK conducted the analysis and drafted the manuscript. LCA, HA, MA, AAlS, EA, AA, ME and IH revised the information presented and suggested edits. All authors revised and approved the final version of the manuscript.

Geolocation information

This study was conducted for Saudi Arabia.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgements

The Authors would like to thank all contributors for their commitment and dedication to this publication. The medical writer support was provided by Syreon Middle east (contracted by Abbvie to run the research project on Economic burden outcomes of atopic dermatitis in Middle East). The authors are fully responsive for all content and editorial decisions were involved at all stages of content development and approved the final version.

Declaration of interest of financial/other interests

Syreon Middle East was a contractual partner of AbbVie BioPharmaceuticals, Inc. ANF, ShA, and ZK are shareholders in Syreon Middle East. BE and EF are employees at Syreon Middle East. MT, HD, SaA, AR, AJ and TH are employees at AbbVie BioPharmaceuticals, Inc. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

ANF, SA, ZK are shareholders in Syreon Middle East. BE and EF are employees at Syreon Middle East. MT, HD, SA, AR, AJ and TH are AbbVie employees and may hold AbbVie stock. For LCA, HA, MAH, AAS, EA, AAT, ME, IH no conflict of interest and no authorship payments were done.

Data availability statement

The authors confirm that the data supporting the findings of this study are available within the article [and/or] its supplementary materials.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

AbbVie funded this research and participated the review and approval of the publication. All authors had access to relevant data and participated in the drafting, review, and approval of this publication. AbbVie sponsored the analysis and interpretation of Data; in reviewing and approval of the final version.