The global burden of stroke and need for a continuum of care

ABSTRACT

Until 4 decades ago, the rates of stroke in low- and middle-income countries were considerably lower than those in more economically robust countries. In the intervening years, however, the rates of stroke in places such as southern India and rural South Africa have approximately doubled, whereas stroke rates in more economically developed nations have decreased. What is far more striking is that rates of disability and mortality arising from stroke are at least 10 times greater in medically underserved regions of the world compared with the most developed nations. The causes of these disparities are clear: above all, there is a lack of primary care treatment to screen patients for stroke risk and to mitigate risk factors. In addition, the lack of access to common drugs and basic medical equipment, as well as the lack of poststroke follow-up programs, rehabilitation, and secondary stroke prevention, means that individuals who would, in countries with better medical care, likely recover from stroke, instead have high rates of death and disability. Several global organizations, most notably the World Health Organization, have formulated and begun to implement public health programs to address these underserved regions. Their success depends on the support and expansion of these efforts so that short-term response to stroke, long-term stroke prevention and care, and screening and treatment of poststroke disabilities can be improved in underserved regions and the human and economic burden on these populations can be minimized.

Stroke is a common and debilitating event associated with a detrimental impact on patients’ health-related quality of life and a high economic cost.^1–5 Although efforts have been made in the United States and other high-income countries to address the causes of stroke and implement appropriate follow-up treatment to reduce stroke-related disability, the situation in low- and middle-income countries has markedly worsened over the last 4 decades.⁶ The primary causes of the increased global burden of stroke are related to the increase in stroke risk factors, particularly hypertension, as well as the underdetection and undertreatment of these risk factors in many developing countries.^7–9

Although there are multiple factors contributing to the increase in stroke risk—such as changes in diet; increasingly sedentary lifestyles; the epidemic of diabetes and obesity, most prominently in high-income countries; and the aging of the population—the barriers to appropriate treatment are also important to consider in stroke management strategies, particularly in developing countries.¹⁰ These barriers include a lack of health care services based in biological sciences for early stroke intervention, as well as a lack of access to a variety of vital resources such as stroke treatment facilities, pharmacologic therapies, and diagnostic equipment—from the simple (e.g., sphygmomanometer) to the more elaborate (e.g., CT and MRI).^10,11 Therapies that might reasonably be expected to be both available and affordable even in low- and middle-income countries are typically neither.¹² Herein, we review the current state of the global burden of stroke, describing key differences between developed and developing countries, discussing proposed interventions to improve the situation, and examining the components of poststroke disability.

GLOBAL BURDEN OF STROKE

According to data from the World Health Organization (WHO), the 2004 global incidence of first-ever stroke was 9.0 million, and 30.7 million people in the world had experienced and survived stroke.⁴ When comparing higher-income countries (defined as those with gross national per capita income [GNPCI] of at least $10,066) with middle- and lower-income countries, one can see a clear difference in stroke trends.⁴ From 1970 through 2008, there was a steady trend toward decreased annual incidence of stroke in higher-income countries, whereas middle- and lower-income countries experienced a remarkable leap in stroke incidence from 1970 through 1989, which then remained high through 2008 (figure 1).⁶ Higher-income countries experienced a combined 42% reduction in stroke incidence compared with an increase of more than 100% in middle- and lower-income countries.⁶ In fact, whereas in the 1970s higher-income countries had a stroke incidence approximately triple that of lower-income countries, the latter group now has a stroke incidence that has surpassed that of the most developed nations.⁶

Figure 1. Age-adjusted stroke incidence in high-income countries vs low- and middle-income countries⁶.

Age-adjusted stroke incidence rates per 100,000 people across the 4 study periods: 1970–1979, 1980–1989, 1990–1999, and 2000–2008. (A) High-income countries. (B) Low- and middle-income countries. Solid line is regression trend line based on a regression of average incidence on study period. Reprinted from The Lancet Neurology, vol. 8, issue 4, Feigin et al.⁶ Worldwide stroke incidence and early case fatality reported in 56 population-based studies: a systematic review, pp. 355–369, ©2009, with permission from Elsevier.

The true impact of stroke must be viewed not only in terms of incidence and mortality rates, but also of disability, which often persists for a long time, and sometimes permanently, after stroke. According to an international survey of 11 middle- and lower-income countries published in 2009 in The Lancet, stroke was the third most common cause of disability among people over the age of 65 years, along with depression.² Nonetheless, although the second leading cause of disability was paralysis or weakness of limbs, the authors acknowledged that this condition may be, to a large degree, a result of stroke. It may be the case that if the contribution of stroke to limb paralysis/weakness could be determined, then that would prove to be the primary cause of disability in this large patient population.² Thus, when placed in the context of the limitations of studies of stroke-related disability in lower-income countries (as is described in more detail below), these data point to a high likelihood that reported rates of stroke, stroke-related mortality, and stroke-related disability may have been underestimated.²

Global burden of stroke: high-income vs middle- and low-income countries

In 2009, Johnston et al.¹³ published a study that compared the burden of stroke across various parts of the world and determined mortality rates and comparable losses of disability-adjusted life years (DALYs) from region to region. In terms of mortality, the differential rates of stroke in various parts of the globe are stark (figure 2). There was as much as a 10-fold difference between the lowest mortality rates, seen primarily in developed nations, and the highest mortality rates, seen primarily in Russia, Eastern Europe, central and northern Asia, and numerous countries across central and western Africa as well as the South Pacific islands. In terms of DALYs, a concept that has gained considerable traction as a means of measuring the burden of disability, even greater differences were seen.¹³ One DALY represents the loss of 1 year of healthy life and is calculated by combining the number of years of life lost to premature mortality with the number of years lost to disability.⁴ Similar to the finding in mortality rates, there was a 10-fold difference in DALY loss between the most affected and the least affected countries, with the highest DALY loss being in Eastern Europe, northern Asia, central Africa, and the South Pacific.¹³

Figure 2. Global stroke mortality rates (adjusted for age and gender)¹³.

Reprinted from The Lancet Neurology, vol. 8, issue 4, Johnston et al.¹³ Global variation in stroke burden and mortality: estimates from monitoring, surveillance, and modeling, pp. 345–354, ©2009, with permission from Elsevier.

Johnston et al.¹³ further observed that a country’s GNPCI significantly predicted stroke mortality rates and loss of DALYs (p < 0.0001 for both). The authors also sought to determine the relationship between GNPCI and stroke risk factors by classifying countries into 4 income groups: low, lower-middle, upper-middle, and upper income. They found that blood pressure levels were nearly equivalent across all 4 groups, whereas increased diabetes risk and total serum cholesterol were both correlated with increased income. The prevalence of tobacco smoking was found to be nearly equivalent in the 3 higher-income groups, which was approximately double that seen in the low-income group. A body mass index ≥25 kg/m² (classified as overweight) was observed in 22% of the low-income countries compared with an ascending rate correlated with income: 48%, 53%, and 59% for the lower-middle, upper-middle, and upper-income countries, respectively.¹³

At first glance, the results of the Johnston study, combined with the data previously described, would seem to be contradictory: lower-income countries with populations possessing fewer stroke risk factors experience considerably higher rates of mortality and DALY loss. A more detailed examination of specific countries and their disposition regarding stroke and its effects may help explain these differences.

New Zealand

New Zealand is a particularly interesting country to study in this context, because it is a highly developed nation that is geographically isolated, with a population consisting of an ethnically European majority and significant indigenous Maori as well as Pacific Islander minorities. Three 1-year studies in Auckland, New Zealand, were conducted a decade apart to determine evolving rates of stroke and changes in stroke risk factors across 2 decades.¹⁴ The first study, conducted in 1981–1982, gathered data from 50% of all the primary care practitioners (PCPs) in the city of Auckland, which had a population of approximately 940,000 according to the 2001 census.¹⁴ The second study, conducted in 1991–1992, gathered all hospital records of stroke, in addition to a cluster sample of 25% of PCPs, in order to capture both hospitalized and nonhospitalized nonfatal strokes. In the third study, conducted in 2002–2003, in addition to gathering hospital records, the investigators attempted to track all new nonhospitalized cases of stroke by contacting every PCP in Auckland. The questionnaire used to assess patient disposition was consistent across the 3 studies. It included a WHO-based definition of first-ever stroke: “rapidly developing clinical signs of focal (or global) disturbance of cerebral function lasting more than 24 hours (unless interrupted by surgery or death) with no apparent cause other than of vascular origin.” Stroke that occurred within 28 days of a previous stroke was defined as “progressing stroke.”¹⁴

Between the first and second studies, the overall incidence of stroke remained stable. The incidence of first-ever stroke, adjusted for age distribution of the world population (per WHO), was 156 per 100,000 people in both the 1981–1982 and 1991–1992 studies, and 139 per 100,000 in the 2002–2003 study.¹⁴ Nearly all of this improvement, however, occurred in the ethnically European population. Among Maoris, there was something of a reverse trend in stroke. Whereas the incidence of stroke for Maoris was approximately 58% lower than that of European New Zealanders in 1981–1982, it had increased substantially by 2002–2003, although it remained approximately 38% lower than that of the European population.¹⁵

Among people of European descent, risk factors in the population shifted in both directions: smoking decreased by approximately half (from 26.7% to 12.6%) among patients with stroke, but obesity and diabetes increased substantially from 1981–1982 to 2002–2003.^14,15 Among Maoris, diabetes and blood pressure remained relatively stable—although higher than in the European population—whereas smoking decreased considerably but remained high, having decreased from 53.3% of stroke patients in 1981–1982 to 38.9% in 2002–2003. The rate of 28-day mortality decreased in both population groups from the first to the third study, from 33% to 21% among Europeans and from 30% to 24% among Maoris, a notably steeper rate of decline being observed among the European population.¹⁵ The ethnic disparities reflected in these studies of the New Zealand population are consistent with studies conducted in the United States and the United Kingdom that demonstrate similar ethnic disparities in stroke risk and incidence.^16–19 For example, in the United States, population studies have also shown a significant racial disparity, with blacks having a higher stroke risk and incidence, with no change over time, in comparison to whites.¹⁷

Southern India

An examination of stroke and stroke risk in southern India provides an opportunity to look at a population with an entirely different socioeconomic history than that of New Zealand, one that has been largely impoverished for many years, but in which certain, relatively small segments have been beneficiaries of recent economic development. Until recently, few epidemiologic data about southern India have been available. The Trivandrum Stroke Registry (TSR) was created to specifically address the limited data regarding stroke in this region.

The TSR study authors evaluated stroke and stroke risk factors in the urban area of Trivandrum (Thiruvananthapuram), which has a population of more than 740,000, and a nearby rural area with a population of nearly 185,000 people (table). Using the WHO World Standard Population, the adjusted incidence of stroke was 135 per 100,000 in the urban community and 138 per 100,000 in the rural community.⁸ Approximately 83% of strokes were ischemic, although this is an estimate because brain imaging was unavailable for 44% of rural patients and 28% of urban patients. The prevalences of risk factors were similar between the rural and urban populations, and in some cases were alarmingly high. Hypertension was seen in 82% and 89% of the urban and rural population, respectively. Diabetes mellitus was present in 49% and 56% of urban and rural study subjects, respectively, and slightly more than a quarter of both populations had dyslipidemia. Smoking was more prevalent in the rural population (39%) compared with the urban population (23%).⁸

Table.

Comparison of stroke type, risk factors, poststroke status, and availability of brain imaging among patients in urban and rural communities in Trivandrum, India^a

The TSR also determined mortality and disability information about patients who had experienced stroke. The 28-day fatality rate was high in the rural population (37%), whereas the urban rate (25%) was closer to that seen in more developed countries. Severe disability was surprisingly higher in the urban population (15% vs 9%), although this difference may be a result of the high rural stroke-related fatality rate. Mild and moderate disabilities were more similar—mild: urban 42% vs rural 43%; moderate: urban 42% vs rural 48%.⁸

Findings of the TSR illustrate the profound disadvantages in stroke survival and outcome in a clinically underserved population. Although stroke rates resemble those of the developed world, mortality in the rural, most underserved population was high. Disability was also high in both the urban and rural populations, and would have been higher still had the mortality rates not been so high.⁸

Rural South Africa

Rural South Africa offers still another unique glimpse at a largely poor population in a country that has been, in many ways, economically robust, but still possesses large impoverished subpopulations.⁹ A study published in 2004 sought to examine the prevalence, rather than the incidence, of stroke in a rural district in the northeastern Limpopo province of South Africa as part of the Southern Africa Stroke Prevention Initiative (SASPI). Compiled in coordination with an annual census, a study population of more than 42,000 individuals aged 15 years or older was identified, and a nonfatal stroke prevalence of 330 per 100,000, adjusted for age per WHO criteria, was determined. This prevalence is lower than that found in most developed countries. At the same time, the prevalence of stroke survivors who required assistance with at least 1 activity of daily living (ADL) was 232 per 100,000, that is, 66% of stroke survivors. This is a very elevated level of disability when compared with the rate of 22% in New Zealand.⁹

The prevalences of stroke risk factors were generally low except for hypertension, which was very high at 71%. It is important to note that the limitations on the investigators were very substantial: they were able to examine only 74% of those reported to have had a stroke, and CT scans were entirely unavailable.⁹ In a separate publication, the SASPI investigators examined the provision of secondary prevention of stroke (i.e., medical resources) among those in the initial study who had experienced stroke. Although nearly all the patients with stroke had sought medical help, the cost of treatment, lack of availability of drugs, aversion to pill taking, delay in receiving treatment, and lack of blood pressure–measuring equipment, when taken together, resulted in very low levels of treatment.¹¹

At least 2 notable observations may be made about the data from the South African study that may also apply to other underserved populations. The first is that despite a low prevalence of stroke, the rate of disability within the stroke-affected population was extremely high, possibly attributable primarily to a lack of available poststroke care. Second, although most risk factors were low in this population, the high rate of hypertension supports the investigators’ suggestion that this rural South African population was (and almost certainly still is) undergoing a transition to increased stroke risk likely to result in a rapid increase in the incidence and prevalence of stroke.⁹ This contention was supported by a subsequent study in the same geographic area several years later, which found, in addition to high rates of hypertension, emerging risk factors for stroke including increased smoking, obesity in women, and subclinical levels of atheroma.²⁰

REDUCTION OF GLOBAL STROKE BURDEN: IMPROVING PREVENTION AND CARE

A variety of different proposals have been presented to improve the stroke burden in middle- and lower-income countries, including those from the WHO and the World Stroke Organization.²¹ In attempting to address this burden, it is important to recognize that developing countries with poor rates of stroke care often possess causative factors different from those in developed countries—such as health care accessibility, resource availability, as well as social and cultural beliefs and expectations—and require targeted interventions to address their particular needs, as opposed to a one-size-fits-all approach.¹⁰ Moreover, treatment of patients with stroke often varies considerably within a given country, the clearest and most common division being between urban and rural areas, as evidenced by the situations in southern India and rural South Africa, where urban dwellers have better access to treatment facilities.¹⁰ For example, time-to-hospital after acute stroke, which is of vital importance if thrombolysis is likely to be effective, has been shown to be similar in locations with hospital access in India and the Philippines compared with that observed in developed countries such as the United Kingdom, although the provision of appropriate thrombolysis is not generally as dependable.^22–24

Regarding feasible efforts toward stroke prevention, proposed public health interventions are aimed primarily at changing the lifestyle variables associated with stroke risk. Such primary prevention strategies implemented by the WHO include promoting a healthy diet—via setting a global maximum salt intake—and physical activity as well as tobacco cessation.¹² Secondary prevention, particularly important in stroke because of the risk of recurrent stroke, is hampered in many developing nations by a lack of even relatively inexpensive basic equipment, the limited availability of drugs, and the fact that treatment is often financially unfeasible for many people who need it.¹⁰ Nonetheless, cost-effective secondary prevention for low- and middle-income populations should include accessible, affordable, and effective antihypertensive and antiplatelet treatments, such as aspirin use for those with high cardiovascular risk, as well as blood pressure and cholesterol control.

Identification of common and emerging risk factors associated with stroke is essential in targeting interventions that could substantially reduce the global burden of stroke. The INTERSTROKE study, published in 2010, evaluated risk factors for first-time acute stroke in 22 countries in order to develop targets for improvement that would be most effective.⁷ Five risk factors were identified that contributed 80% of stroke risk: hypertension, smoking status, waist-to-hip ratio, diet risk score (based around a healthy cardiovascular diet), and physical activity.⁷ Five other risk factors accounted for an additional 10% of stroke risk: diabetes mellitus, alcohol use, psychosocial factors (e.g., stress, depression), cardiac causes (i.e., atrial fibrillation, previous myocardial infarction, valvular heart disease), and ratio of apolipoprotein B to apolipoprotein A1.⁷

In addition to pathophysiologic risk factors for stroke, external and infrastructural factors have a large role in stroke risk; this is why equitable stroke-related services are central to improving the stroke burden in underserved areas. In a recent publication, the WHO delineated the key components of an equitable approach to stroke services. They include: 1) primary care assessment and management of cardiovascular risk and TIA; 2) secondary prevention measures; 3) education of the public and health care providers about stroke prevention and management; 4) access to stroke care and rehabilitation services; and 5) community and family support for patients with stroke and their caregivers.¹² The WHO has suggested—and implemented in several cases—a stepwise approach to stroke surveillance using standardized tools for data collection. This phased implementation of interventions is cause for some optimism, provided that there is an urgent, stronger response to stroke and other chronic diseases in middle- and lower-income countries.²¹

SPECTRUM OF POSTSTROKE DISABILITIES AND GLOBAL IMPACT

Disabilities resulting from stroke differ much among patients but are frequently profound. Common disabilities include limb weakness, spasticity, pain, psychosocial issues (including depression, cognitive decline, and social isolation), language impairments, and functional limitations including problems with walking, standing, and inability to perform ADLs. Measurements of the prevalence of spasticity poststroke are somewhat variable, based on the criteria used and the patient population measured. That said, between 20% and 40% of patients who have had a stroke are likely to experience spasticity 3 to 12 months after the index event.^25–30 A Swedish study examining quality-of-life–related issues including manual dexterity, mobility, walking/falling, and performance of ADLs reported significant reductions in every one of these domains among patients with poststroke spasticity.³⁰ Another Swedish study reported a prevalence of 17% for spasticity and 4% for disabling spasticity 1 year after stroke.²⁹ Moreover, language impairments are seen in approximately one-quarter to one-third of first-time patients with stroke.^31,32 Aphasia creates obvious difficulties for poststroke patients experiencing it, and also increases their dependence on caregivers. Greater severity of aphasia also increases both short-term (1-month) and long-term (10-year) mortality risk.³²

Regarding the global impact of poststroke disability, a recently published study by Ferri et al.³³ examined survivors of stroke in several Latin American countries as well as in India and China. Severe disability (per the WHO Disability Assessment Schedule) varied from region to region, with a rate of 18.5% in rural Mexico compared with 72.5% in rural India.³³ Inability to stand or severe difficulty in standing for 30 minutes was observed in 13.3% of rural Peruvians and 63.6% of rural Indians, and inability to walk or severe difficulty in walking 1 km was reported for 20% of rural Peruvians and 63.6% of rural Indians. Both dementia and depression were highly correlated with severe disability, and were observed in 25.4% and 10.8% of the overall study population, respectively.³³ These data are consistent with a study from Poland—a country characterized by high stroke rates—which found that approximately one-third of patients with stroke had depression, and that depression was associated with worse functional status, greater difficulties with ADLs, and a pattern of self-imposed social isolation.³⁴ Thus, severe poststroke disability and dependence are common and have been shown to be comorbid with physical impairments, dementia, and depression. Additionally, the Ferri et al.³³ study showed that one-quarter of caregivers for stroke survivors had to cut back on work to provide care. Furthermore, in addition to stroke prevention, the minimal support from social and health services for rehabilitation and long-term care for stroke survivors in these underserved regions is also a cause for concern.

CONCLUSIONS

Stroke is a devastating event for individuals in any country, but for patients in less-developed regions of the world, stroke is frequently associated with a greater risk of mortality and disability. Although the incidence of stroke was lower in developing nations than in developed nations in the past, this is no longer the case. The rates of poststroke mortality and morbidity are considerably higher in many lower-income regions compared with high-income industrialized nations. These differences can largely be attributed to a lack of primary care treatment that would identify and mitigate stroke risk, poststroke follow-up programs, rehabilitation, and secondary stroke prevention. A lack of basic equipment and drugs necessary for the prevention and treatment of stroke is also a key factor. Improving this situation requires the implementation of existing public health programs developed by the WHO, among others, that would provide long-term follow-up services for stroke prevention as well as screening and treatment of poststroke disabilities.

Glossary

ADL

activity of daily living

DALYs

disability-adjusted life years

GNPCI

gross national per capita income

PCP

primary care practitioner

SASPI

Southern Africa Stroke Prevention Initiative

TSR

Trivandrum Stroke Registry

WHO

World Health Organization

DISCLOSURE

Dr. Norrving served on clinical trial steering, end-point adjudication, and data monitoring committees for Servier, PhotoThera, and Sygnis Pharma AG, and on scientific advisory boards for Allergan, Inc., and Bayer HealthCare Pharmaceuticals. He receives government funding from the European Union European Implementation Score (EIS) project. He also receives publishing royalties for contribution to the book Hjärnan, published by Karolinska Institutet University Press, 2007. Dr. Kissela served on a scientific advisory board for and received funding for travel and speaker honoraria from Allergan, Inc. He received research support from NexStim and grant funding from NIH/National Institute of Neurological Disorders and Stroke. He also provides medicolegal case reviews.