The impact of seven major noncommunicable diseases on direct medical costs, absenteeism, and presenteeism in Gulf Cooperation Council countries

Abstract

Aims

To estimate the current burden of seven major noncommunicable diseases on direct medical costs, absenteeism, and presenteeism in the six countries in the Gulf Cooperation Council: Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates.

Materials and methods

We used data from pre-existing datasets and the literature. We identified seven major noncommunicable diseases for which data were available: coronary heart disease, stroke, type-2 diabetes mellitus, breast cancer, colon cancer, chronic obstructive pulmonary disease, and asthma. We estimated the per unit cost (the annual cost of treating each illness for one person) of each disease, multiplied per unit cost by disease prevalence counts to generate disease-specific costs, and then summed across diseases. We calculated the cost of absenteeism and presenteeism by multiplying the gross domestic product per person in the labor force by the loss in productivity from each disease due to absenteeism and presenteeism, respectively, and the prevalence in the labor force of each disease.

Results

We estimate that the direct medical costs of seven major noncommunicable diseases in Gulf Cooperation Council countries are $16.7 billion (2019 International $), equal to 0.6% of gross domestic product. We estimate that absenteeism and presenteeism due to these seven noncommunicable diseases cost 0.5 and 2.2% of gross domestic product, respectively.

Limitations

Our study does not capture all noncommunicable diseases and does not capture all types of indirect costs. Our cost estimates are particularly sensitive to our assumptions regarding type-2 diabetes mellitus.

Conclusion

The economic burden of noncommunicable diseases in Gulf Cooperation Council countries is substantial, suggesting that successful preventive interventions have the potential to improve both population health and reduce costs. Further research is needed to capture a broader array of noncommunicable diseases and to develop more precise estimates.

Keywords:

• Gulf Cooperation Council
• noncommunicable diseases
• direct medical costs
• absenteeism
• presenteeism

JEL Classification codes:

• I10
• I1
• I
• I19

Introduction

Much has been written about the economic costs of noncommunicable diseases (NCDs) globally^1,2. NCDs not only increase the direct costs of health care, but also impose indirect costs such as increased absenteeism (lost output due to missed days of work) and presenteeism (lost output due to diminished productivity while at work). Quantifying these costs draws attention to the additional burden that NCDs impose, beyond traditional measures of morbidity and mortality. These costs are thus useful in planning and resource allocation decisions.

Relatively few studies have attempted to quantify economic costs caused by NCDs in the countries that comprise the Gulf Cooperation Council (GCC), an intergovernmental political and economic union that consists of all the Arab countries of the Persian Gulf region other than Iraq. To date, most of the few studies that have been conducted focused on one condition in one country. Salman, AlSayyad, and Ludwig (2019), for example, estimated the direct cost of type-2 diabetes mellitus in Bahrain³. Al-Busaidi, Habibullah, and Soriano (2013) estimated the direct cost of asthma in Oman⁴. Khadadah (2013) estimated the direct cost of asthma in Kuwait⁵. Al-Maskari (2010) estimated the direct cost of type-2 diabetes mellitus among patients without complications in United Arab Emirates⁶. Alzaabi, Alseiari, and Mahboub (2014) estimated the direct cost of treating asthma patients in Abu Dhabi, United Arab Emirates⁷. These studies are not comparable, however, because they relied on different methods and data sources. Moreover, these studies estimated direct medical costs only, omitting indirect costs entirely.

Given the paucity of evidence on the economic burden of NCDs in GCC countries, the objective of this study was to provide estimates of direct and indirect costs in the six GCC countries: Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and United Arab Emirates. We estimated three types of costs that arise from seven major NCDs: (1) direct medical costs, including the cost of hospitalizations, outpatient visits, emergency department visits, general practitioner visits, and prescription drugs; and indirect costs arising from (2) absenteeism and (3) presenteeism.

Methods

Our analysis relied upon pre-existing datasets, as described below. We used a bottom-up approach to estimate costs. We started by identifying seven NCDs for which data were available: (1) coronary heart disease, (2) stroke, (3) type-2 diabetes mellitus, (4) breast cancer, (5) colon cancer, (6) chronic obstructive pulmonary disease (COPD), and (7) asthma. We excluded many NCDs for which data were not available such as major depression, anxiety disorders, Alzheimer’s, epilepsy, sleep disorders, and lung cancer.

To estimate direct medical costs, we began by estimating the prevalence and per unit cost—the annual cost of treating each illness for one person—for each of the seven NCDs separately in each of the six GCC countries. We used prevalence data from the Institute for Health Metrics and Evaluation’s Global Disease Burden database⁸. Our per unit cost estimates for Saudi Arabia, Kuwait, Qatar, and United Arab Emirates for coronary heart disease, stroke, type-2 diabetes mellitus, breast cancer, and colon cancer are from Ding, Lawson, and Kolbe-Alexander et al. (2016)⁹. Since this study did not provide per unit cost estimates for Bahrain and Oman, we imputed per unit cost data for these two countries based on the median per unit cost in the other four GCC countries (i.e. the mean of the per unit costs in the middle two countries). We then multiplied this estimate by the ratio of per capita health spending in Bahrain or Oman, respectively, to median per capita health spending in the other four GCC countries to account for differences in overall health spending across countries.

Because Ding, Lawson, and Kolbe-Alexander et al. (2016)⁹ did not provide per unit cost estimates for COPD and asthma, we obtained estimates from the literature. Our per unit cost estimates for COPD were based on the per unit cost reported in a German study¹⁰, adjusting for overall per capita health spending in each GCC country relative to Germany. We chose this study because the estimated per unit cost was the median among three studies^10–12 identified in the literature. Our per unit cost estimates for asthma were based on estimates from Abu Dhabi, United Arab Emirates⁷. As above, we adjusted these estimates by each country’s per capita health spending relative to per capita health spending in the United Arab Emirates.

We updated monetary figures in local currencies to 2019 costs using country-specific annual inflation rates. We then converted these figures to 2019 International dollars ($) by dividing local currency by the Purchasing Power Parity exchange rate. All our cost estimates are reported in 2019 International $. We multiplied the prevalence rate for each disease in each country by the population in each country¹³ to obtain estimates of the prevalence (number of cases) of each disease in each country, then multiplied the number of cases by per unit annual costs to arrive at an estimate of total annual direct medical costs for each condition in each country. We summed up the costs of each disease in each country to obtain estimates of total annual direct medical costs for all seven NCDs in each country.

We used the same estimates as Rasmussen, Sweeny, and Sheehan (2016)¹⁴ regarding the percentage productivity loss due to absenteeism and presenteeism per employee per year for each of the seven diseases in our analysis. These estimates were based upon earlier research by Goetzel, Long, and Ozminkowski et al. (2004)¹⁵ on productivity loss due to absenteeism and presenteeism by disease. We multiplied the loss of productivity due to absenteeism or presenteeism by the estimated number of cases of each disease among part- and full-time workers and by per capita gross domestic product (GDP) among those in the workforce to generate disease-specific absenteeism and presenteeism costs. To obtain estimates of total productivity losses, we summed absenteeism and presenteeism costs. We summed across diseases to generate total absenteeism, presenteeism, and productivity losses. For ease of comparison across countries, we compared direct medical costs to GDP in each GCC country. We also compared absenteeism, presenteeism, and productivity losses to each country’s GDP. These comparisons provide perspective of the burden of NCDs across GCC countries.

To test the variation of our results to our assumptions, we conducted several sensitivity analyses. We considered the impact of replacing our base case type-2 diabetes mellitus per unit cost estimates⁹ with lower and higher per unit cost estimates from published studies conducted in Saudi Arabia^16,17, adjusting these estimates upward or downward based on per capita health spending in each GCC country relative to Saudi Arabia. We also assessed the impact of imputing per unit COPD costs based on estimates from Greece¹¹ or the United States¹², respectively, rather than from the German study used in our base case analysis. Again, we adjusted these estimates based on per capita health spending in each country relative to the United States or Germany, respectively. We replaced type 2 diabetes mellitus prevalence data from the GBD database⁸ with higher prevalence estimates from the International Diabetes Foundation¹⁸. Finally, we replaced our estimates of the impact of stroke and type-2 diabetes mellitus on absenteeism and presenteeism with estimates used in a recent unpublished Kuwait study¹⁹. These estimates were derived from four previously published studies: Salman, Alsayyad, and Ludwig³, Mitchell and Bates (2011)²⁰, Wang, Beck, and Berglund et al. (2003)²¹, and Bommer, Heesemann, and Sagalova et al.²².

Results

Estimated mortality, prevalence, and per capita annual medical costs for each condition are presented in Table 1. In all six GCC countries, coronary heart disease causes the most deaths of the seven NCDs considered here; type-2 diabetes mellitus is the most prevalent, with estimates ranging from 4.4% in Oman to 13.3% in Bahrain. Following type-2 diabetes mellitus are asthma, with estimates ranging from 2.5% in Saudi Arabia to 6.8% in the United Arab Emirates, and coronary heart disease, where the lowest estimated prevalence is in Qatar (1.7%) and the highest is in Bahrain (3.5%). Breast and colon cancer are relatively uncommon in all countries, with prevalence rates no higher than 0.2% (breast cancer prevalence in Kuwait and Bahrain). Estimated per capita annual medical costs range from a low of $37 for asthma in Oman to a high of $4,569 for colon cancer in Qatar.

Table 1. Estimated mortality, prevalence, and annual per unit cost of NCDs in GCC countries, 2019.

	Coronary heart disease^b	Stroke^b	Type 2 diabetes mellitus^b	Breast cancer^b	Colon cancer^b	COPD^c	Asthma^d
Estimated mortality rate (deaths per 100,000 population)^a
Bahrain	59.8	17.0	48.5	7.2	5.8	7.7	2.2
Kuwait	58.7	20.8	8.0	3.9	5.0	3.4	0.7
Oman	74.4	22.5	14.0	2.9	2.9	4.4	0.2
Qatar	29.0	6.3	11.7	3.1	2.9	.7	0.3
Saudi Arabia	83.1	35.5	7.8	4.0	4.6	5.9	2.5
UAE	52.8	23.5	13.3	4.9	4.9	11.8	6.0
Estimated prevalence^a
Bahrain (%)	3.5	1.0	13.3	0.2	0.1	1.4	3.3
Kuwait (%)	3.2	1.0	8.9	0.2	0.1	0.9	4.1
Oman (%)	2.1	0.9	4.4	0.1	0.0	0.8	3.9
Qatar (%)	1.7	0.8	10.0	0.1	0.0	1.0	3.1
Saudi Arabia (%)	2.4	1.4	7.2	0.1	0.1	1.3	2.5
UAE (%)	2.2	1.5	8.0	0.1	0.0	2.1	6.8
Estimated per unit annual direct medical cost (2019 International $)
Bahrain	1,264	3,077	2,644	1,176	3,504	2,648	359
Kuwait	1,496	3,640	3,239	1,391	4,145	3,306	448
Oman	130	316	271	121	359	272	37
Qatar	1,648	4,013	3,071	1,534	4,569	3,015	409
Saudi Arabia	958	2,331	1,936	891	2,655	3,056	414
UAE	1,406	3,422	3,456	1,308	3,898	2,772	376

Abbreviations. NCD, noncommunicable disease; UAE, United Arab Emirates; COPD, chronic obstructive pulmonary disease; estimates may not sum to totals due to rounding.

^aIHME 2020⁸.

^bKuwait, Qatar, Saudi Arabia, and UAE estimates are from Ding, Lawson, Kolbe-Alexander et al. 2016⁹; Bahrain and Oman estimates are imputed based on the median per capita cost estimate in Kuwait, Qatar, Saudi Arabia, and UAE.

^cCOPD estimates are based on Wacker, Jörres, Schulz et al. (2016)¹⁰.

^dAsthma estimates are based on Alzaabi, Alseiari, Mahboub (2014)⁷.

Multiplying these estimates by the number of individuals in each country with each condition and summing across conditions reveals that the annual total direct medical costs of these seven NCDs in the six GCC countries is $16.7 billion. Of this total, 61.2% is attributable to type-2 diabetes mellitus, 13.0% is due to COPD, 11.6% is due to stroke, and 8.8% is due to coronary heart disease. Only 0.5% is due to breast cancer and 0.5% is due to colon cancer. Table 2 provides a breakdown of estimated annual direct medical costs for each condition in each country.

Table 2. Estimated annual direct medical costs due to NCDs in GCC countries.

	Estimated annual direct medical costs (millions of 2019 Int $)
	Coronary heart disease	Stroke	Type 2 diabetes mellitus	Breast cancer	Colon cancer	COPD	Asthma	Total
Bahrain	73	53	578	4	3	63	19	794
Kuwait	204	156	1,218	9	10	129	78	1,806
Oman	13	13	59	1	1	10	7	104
Qatar	79	93	871	6	5	82	36	1,172
Saudi Arabia	798	1,118	4,796	43	45	1,320	352	8,472
UAE	302	508	2,701	15	15	577	251	4,371
Total	1,469	1,942	10,224	78	80	2,181	744	16,719

Abbreviations. NCD, noncommunicable disease; UAE, United Arab Emirates; COPD, chronic obstructive pulmonary disease; estimates may not sum to totals due to rounding.

The amount that each GCC country spends on these NCDs relative to its GDP is shown in Table 3. Relative to GDP, Oman and Qatar spend the least on these NCDs (0.4% of GDP) and Bahrain spends the most (1.0% of GDP).

Table 3. Estimated total direct medical costs vis-à-vis GDP in GCC countries.

	Total direct medical costs (millions of 2019 International $)	Total direct medical costs as a%age of GDP (%)
Bahrain	794	1.0
Kuwait	1,806	0.8
Oman	104	0.4
Qatar	1,172	0.4
Saudi Arabia	8,472	0.5
UAE	4,371	0.6
GCC	16,719	0.6

Abbreviations. NCD, noncommunicable disease; UAE, United Arab Emirates; GCC, Gulf Cooperation Council.

Estimated absenteeism, presenteeism, and productivity costs are presented in Table 4. Across the six GCC countries, estimated absenteeism and presenteeism costs due to NCDs total $15.3 billion and $65.3 billion, respectively. Combining these two costs reveals that total productivity losses are $80.6 billion annually. The largest driver of absenteeism costs is asthma. Type-2 diabetes mellitus is the largest driver of presenteeism costs. By contrast, breast cancer and colon cancer—which are much less prevalent than type-2 diabetes mellitus and asthma—have relatively minor effects on absenteeism and presenteeism costs.

Table 4. Estimated absenteeism, presenteeism, and productivity costs Due to NCDs in GCC countries.

	Coronary heart disease	Stroke	Type 2 diabetes	Breast cancer	Colon cancer	COPD	Asthma	Total
Estimated annual absenteeism costs (millions of 2019 Int $)
Bahrain	94	27	101	15	4	82	107	430
Kuwait	253	78	199	32	12	154	341	1,068
Oman	109	43	65	18	4	84	217	534
Qatar	150	72	254	29	9	182	333	1,027
Saudi Arabia	1,518	855	1,290	219	78	1,688	1,857	7,505
UAE	493	338	513	67	22	1,032	2,284	4,749
Total	2,616	1,413	2,422	372	129	3,223	5,140	15,314
Estimated annual presenteeism costs (millions of 2019 Int $)
Bahrain	227	66	1,438	18	5	232	236	2,223
Kuwait	614	190	2,838	38	14	435	751	4,879
Oman	265	106	928	14	5	237	478	2,033
Qatar	363	174	3,613	35	11	513	732	5,441
Saudi Arabia	3,686	2,077	18,386	265	95	4,761	4,085	33,354
UAE	1,198	820	7,303	82	27	2,910	5,025	17,365
Total	6,353	3,432	34,506	452	156	9,087	11,307	65,295
Estimated annual productivity losses (millions of 2019 Int $)
Bahrain	321	93	1,540	32	9	314	344	2,653
Kuwait	866	268	3,037	69	26	589	1,092	5,947
Oman	374	149	993	26	9	321	696	2,567
Qatar	513	245	3,866	64	20	696	1,064	6,468
Saudi Arabia	5,204	2,932	19,676	484	173	6,449	5,942	40,859
UAE	1,691	1,158	7,816	149	50	3,942	7,309	22,115
Total	8,969	4,846	36,928	824	285	12,310	16,447	80,609

Abbreviations. NCD, noncommunicable disease; UAE, United Arab Emirates; COPD, chronic obstructive pulmonary disease; productivity losses, absenteeism costs + presenteeism costs; estimates may not sum to totals due to rounding.

Estimated absenteeism, presenteeism, and total reduced productivity costs as percentages of GDP are shown in Table 5. Across all six GCC countries, absenteeism and presenteeism costs are equal to 0.5% of GDP and 2.2% of GDP, respectively. The cost of absenteeism is highest in Oman (1.8% of GDP) and lowest in Qatar and Saudi Arabia (0.4% of GDP). The cost of presenteeism is highest in Oman (6.9% of GDP) and lowest in Qatar and Saudi Arabia (2.0% of GDP). Summing absenteeism and presenteeism costs reveals that the highest cost resulting from productivity losses is in Oman (8.7% of GDP) and the lowest is in Qatar and Saudi Arabia (2.4% of GDP).

Table 5. Estimated absenteeism, presenteeism, and productivity costs vis-à-vis GDP in GCC countries.

	Total costs (millions of 2019 international $)			Total costs as a % of GDP
	Absenteeism	Presenteeism	Productivity	Absenteeism (%)	Presenteeism (%)	Productivity (%)
Bahrain	430	2,223	2,653	0.6	2.9	3.4
Kuwait	1,068	4,879	5,947	0.5	2.2	2.7
Oman	534	2,033	2,567	1.8	6.9	8.7
Qatar	1,027	5,441	6,468	0.4	2.0	2.4
Saudi Arabia	7,504	33,354	40,859	0.4	2.0	2.4
UAE	4,749	17,365	22,115	0.7	2.5	3.2
GCC	15,314	65,295	80,609	0.5	2.2	2.7

Abbreviations. UAE, United Arab Emirates; GCC, Gulf Cooperation Council. Sums may not add to totals due to rounding.

In the above analyses, we relied on country-specific per unit type-2 diabetes mellitus costs from Kuwait, Qatar, Saudi Arabia, and United Arab Emirates reported by Ding, Lawson, Kolbe-Alexander et al.⁹. These estimates ranged from a low of $271 in Oman to a high of $3,456 in the United Arab Emirates. Replacing these with much higher estimates based on Mokdad, Tuffaha, Hanlon et al. (2015)¹⁶—who reported per unit costs of $9,009 in Saudi Arabia—increases combined direct medical costs in the six GCC countries more than twofold to $43.3 billion (1.5% of GDP). Replacing per unit type-2 diabetes mellitus costs with a lower estimate from Almutairi and Alkharfy (2013)¹⁷—who reported per unit costs of $1,605—reduces combined direct medical costs to $13.0 billion (0.4% of GDP). Replacing per unit COPD costs with higher or lower estimates from the literature^11,12 has relatively little impact on total direct medical costs. Replacing our base case type 2 diabetes mellitus prevalence estimates⁸ with higher estimates from the International Diabetes Foundation¹⁸ increases estimated direct medical costs, absenteeism costs, and presenteeism costs to $22.9 billion (0.8% of GDP), $16.8 billion (0.6% of GDP) and $86.1 billion (2.9% of GDP), respectively. Replacing our base case estimates of the effect of stroke and type 2 diabetes mellitus on absenteeism and presenteeism with estimates from a Kuwait study¹⁹ has virtually no effect on absenteeism but reduces presenteeism costs by roughly half (from $65.3 billion to $30.7 billion). A summary of our sensitivity analyses is provided in Table 6.

Table 6. Sensitivity analyses.

	Direct medical costs		Absenteeism costs		Presenteeism costs		Productivity costs
	Int $	% of GDP	Int $	% of GDP	Int $	% of GDP	Int $	% of GDP
Base case	16.7	0.6	15.3	0.5	65.3	2.2	80.6	2.7
Higher diabetes per unit cost^a	43.3	1.5	15.3*	0.5*	65.3*	2.2*	80.6*	2.7*
Lower diabetes per unit cost^b	13.0	0.4	15.3*	0.5*	65.3*	2.2*	80.6*	2.7*
Higher COPD per unit cost^c	18.3	0.6	15.3*	0.5*	65.3*	2.2*	80.6*	2.7*
Lower COPD per unit cost^d	15.3	0.5	15.3*	0.5*	65.3*	2.2*	80.6*	2.7*
Higher diabetes prevalence^e	22.9	0.8	16.8	0.6	86.1	2.9	102.9	3.5
Alternative assumptions about the impact of stroke and type 2 diabetes mellitus on absenteeism and presenteeism	16.7*	0.6*	15.6	0.5	30.7	1.0	46.3	1.6

All figures refer to the Gulf Cooperation Council countries in the aggregate; Int $, billions of international dollars; an asterisk (*) indicates no change from the base case; productivity loss absenteeism costs + presenteeism costs; absenteeism and presenteeism costs may not sum to productivity loss due to rounding.

^aMokdad, Tuffaha, Hanlon et al. (2015)¹⁶.

^bAlmutairi and Alkharfy (2013)¹⁷.

^cDalal, Christensen, Liu, Riedel (2010)¹².

^dSouliotis, Kousoulakou, Hillas et al. (2017)¹¹.

^eInternational Diabetes Foundation (2019)¹⁸.

Discussion

These estimates reveal the substantial toll that seven major NCDs impose in the GCC. Despite limiting our analysis to a small number of NCDs, estimated annual direct medical costs total $16.8 billion, which represents 0.6% of GDP in these countries. The estimated impact of NCDs on workforce productivity is even higher: We estimate that the seven NCDs we examined raise annual absenteeism and presenteeism costs by $15.3 billion (0.5% of GDP) and $65.3 billion (2.2% of GDP), respectively.

Direct costs

There is a large literature on the direct and indirect cost of NCDs in Western countries^23–29, but few comparable studies in GCC countries. This comparison suggests our estimates are likely to be conservative. The UN Interagency Task Force on NCDs³⁰ used National Health Accounts data to estimate that the direct medical costs of cardiovascular diseases, cancers, endocrine and metabolic diseases, and respiratory diseases accounted for 0.8% of GDP in Saudi Arabia. By comparison, estimated direct medical costs in Saudi Arabia in our study were 0.5% of GDP.

A Kuwait study conducted by Ministry of Health Kuwait et al. (2020)¹⁹ estimated that the direct medical costs of four NCD categories—cardiovascular disease, diabetes, cancer, and respiratory illnesses—were equal to 37.3% of total health spending in 2018. By comparison, we estimate that the seven NCDs we examined led to direct medical costs equal to just 15.6% of health spending in Kuwait in 2019. These two analyses, however, are not comparable because of differences in the way the disease groups are defined. For example, we limited our analysis to Kuwaitis who had either breast or colon cancer (9,256 individuals, assuming no overlap), whereas the Kuwait analysis considered Kuwaitis with any form of cancer (30,999 individuals). Our analysis included Kuwaiti patients with COPD or asthma (213,299 individuals, assuming no overlap), whereas the Kuwait study examined individuals with any form of chronic respiratory disease (469,561 individuals). Had we considered broader disease categories—as the Kuwaiti authors did—our estimates would have been closer to those reported in the Kuwait study.

Previous studies have reported highly divergent estimates of the per unit cost of type-2 diabetes mellitus^9,16,17, the largest driver of direct medical costs in our analyses. We took this uncertainty into account by performing sensitivity analyses using a range of estimates drawn from published sources. Our results are very sensitive to our assumptions regarding the per unit cost of type-2 diabetes mellitus. Our assumptions regarding type-2 diabetes mellitus prevalence also affected our results significantly, albeit to a smaller extent than our assumptions regarding the per unit cost. The reason for this is that type-2 diabetes mellitus prevalence estimates do not differ by as much as the per unit cost estimates.

Changing our estimate of the per unit cost of COPD had relatively little effect on our results because COPD is responsible for a relatively small proportion of NCD direct costs.

Indirect costs

Absenteeism and presenteeism costs resulting from the seven NCDs in Saudi Arabia are equal to 0.4 and 2.0% of GDP, respectively. By comparison, Rasmussen, Sweeny, and Sheehan¹⁴ estimated absenteeism and presenteeism costs due to NDCs in Saudi Arabia of 1.5 and 4.3% of GDP, respectively. The discrepancies between our estimates and theirs are due to our more limited list of NCDs; our estimates would be much higher if we were to include a larger number of NCDs in our analysis. No other comparable estimates in GCC countries were available.

As noted above, our results show that asthma is the largest driver of absenteeism costs whereas type-2 diabetes mellitus is the largest driver of presenteeism costs. There is some evidence that treatment of asthma has a greater beneficial effect on presenteeism than on absenteeism (Sadatsafavi, Rousseau, and Chen et al. 2014)³¹. It is possible that the inverse is true for type-2 diabetes mellitus, which (for various physiological reasons) often is associated with increased urination. In a Japanese survey of workers who had been treated for type-2 diabetes mellitus, 10.5% of respondents said they had to use the toilet frequently and that this adversely affected their work (Nakajima 2017)³². It is possible that other type 2 diabetes mellitus complications (e.g. sensitivity to light and neuropathy) and effects of poorly controlled blood sugar levels (e.g. diminished mental clarity) also adversely affect presenteeism to a greater extent than absenteeism.

On the other hand, it is also possible that our estimate of the impact of type 2 diabetes mellitus on presenteeism—drawn from Rasmussen, Sweeny, and Sheehan¹⁴—is too high. Replacing our estimates with those used by the authors of the aforementioned Kuwait study¹⁹ sharply reduces the effect of type 2 diabetes mellitus on presenteeism costs.

Oman’s high productivity losses vis-à-vis other GCC countries (Table 5) reflect its large labor force (third-largest in the GCC behind Saudi Arabia and the United Arab Emirates) relative to the size of its economy (smallest GDP by far in the GCC). In other words, Oman’s economy is more labor-intensive than those of its GCC neighbors, so it incurs greater absenteeism and presenteeism costs when people are ill.

Conclusion

Overall, our base case analysis of direct medical costs of the included NCDs is conservative due to our use of type-2 diabetes mellitus per unit cost and prevalence estimates^8,9 that are on the low end of published estimates. Moreover, as noted above, our analysis included only seven out of dozens of NCDs.

Our analysis of indirect costs, by contrast, may be either downwardly or upwardly biased: Downward bias results from the limited number of NCDs in our analysis, whereas upward bias may result from our use of a high-end estimate of the effect of type 2 diabetes mellitus on presenteeism.

The indirect costs of NCDs are not limited to the costs arising from absenteeism and presenteeism. Incorporating indirect costs due to premature mortality, reduced labor force participation, care provided by family and friends, and intangible costs such as pain and suffering would add to these totals. Such costs were omitted from our analysis.

Although it was not possible to catalogue all of the costs of NCDs—and some costs were not calculated with precision due to uncertainties in the underlying parameters—our results nevertheless indicate that NCDs impose a significant economic burden on GCC countries. This raises the possibility that interventions that prevent NCDs can simultaneously improve health outcomes and reduce economic costs. Since the median age in GCC countries is 28.4 years¹³ and those at highest risk of most NCDs are middle-aged and elderly adults, the economic burden of NCDs is likely to increase in the future in the absence of interventions, providing further financial justification for implementing programs aimed to reduce risk factors for NCDs.

Our analysis shows that the most significant driver of NCD costs is type-2 diabetes mellitus, which is heavily influenced by lifestyle factors—in particular diet and exercise choices that lead to obesity. Type-2 diabetes also has the potential to increase costs for the other included NCDs. Standard treatment typically includes healthy eating, frequent exercise, and weight loss, which would also reduce risk factors for many other NCDs. Diabetes treatment may also include diabetes medication or insulin therapy along with blood sugar monitoring. Preventive interventions that reduce obesity (and hence type-2 diabetes mellitus) should be strongly considered. This includes the interventions designated as “best buys” by the World Health Organization for prevention and treatment of NCDs³³ and other cost-effective interventions such as Saudi Arabia’s excise tax on sugar-sweetened beverages^34,35.

Transparency

Declaration of funding

Funding for this research was provided by the World Bank under its advisory services program to the Gulf Cooperation Council countries. The sponsor—the World Bank—participated in the preparation of this paper.

Declaration of financial/other interests

No potential conflict of interest was reported by the author.

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

Author contributions

EAF, JM, CH, and SES conceived, designed, and coordinated the assessment. EAF and JM analyzed the data. All authors participated in drafting the manuscript. All authors read and agreed to publish the manuscript.

Acknowledgements

Open Access funding provided by the Qatar National Library. The authors are thankful to Issam Abousleiman, World Bank Country Director for GCC countries; Rekha Menon, Practice Manager for Health Nutrition and Population in the MENA region; and Keiko Meiwa, Regional Director for Human Development, World Bank, for the support throughout. The authors are also thankful to Wael Ahmed Shelpai and Hira Abdulrazzaq from the United Arab Emirates who provided critical technical inputs into the draft.