The Trend of Healthcare Needs among Elders and Its Association with Healthcare Access and Quality in Low-Income Countries: An Exploration of the Global Burden of Disease Study 2019

To investigate the trend of healthcare needs among elders in low-income countries (LICs) and how changes in healthcare access and quality (HAQ) have correlated with these changes from 1990 to 2019, this study used estimates from the global burden of disease (GBD) 2019 study, including prevalence, years of life lost (YLLs), years lived with disability (YLDs), life expectancy (LE), health-adjusted life expectancy (HALE) and the HAQ index for years 1990 and 2019. We found increases in numbers of YLLs, YLDs, and prevalent cases due to NCDs, and the rate of increase was higher for all indicators of non-communicable diseases (NCDs) when compared with communicable, maternal, neonatal and nutritional diseases among elders. We also observed increases in LE and HALE among all countries. However, this was also challenged by increases in unhealthy life years (ULYs) and their constant percentage of LE. The HAQ index of LICs was also found to be low, although it had increased during the period. A reduction in the burden of acute diseases explains the increase in LE, but increases in ULYs and the NCD burden were also observed. LICs need to improve their HAQ to counter the growing threat of longer but less healthy lives.


Introduction
Aging is one of the biggest concerns of our time around the globe. This is simply because many developed nations have experienced population transitions into aged societies, and the World Health Organization (WHO) projects that by 2050, one in every five people will be over 60 years old [1]. As the focus on aging continues to be in high-income countries (HICs), low-income countries (LICs) (countries with a gross national income per capita less than or equal to 1085 US dollars in 2021) [2] usually feel unbothered, but indeed the populations in these countries too are aging. Nonetheless, numerous studies conducted in aging societies indicate the need for shifts in healthcare provision and policies to match the increased need for healthcare that comes with increases in aged populations [3][4][5].
The population of elderly people was projected to be around 40 million people in lowincome countries in 2022, but this is expected to increase to 100 million people in 2050 [6]. This increase in the number of older people is easily attributable to longer life expectancy, which is a result of the successful minimization of premature mortality. In their study, which included 54 low-income countries, Hauck et al. noted that life expectancy increases were associated with a reduction in the human immunodeficiency virus prevalence among

Data Sources
The data used in this study was extracted from the Institute of Health Metrics and Evaluation's (IHME) Global Burden of Disease (GBD) 2019 database. The GBD 2019 study provides data that is used to measure countries' health challenges and how they vary over time and among countries, regions and subregions, among other classifications. The GBD 2019 study included the categorization of 369 diseases and injuries and attributed burden to 87 risk factors from 204 countries and territories categorized by sex and age groups for indicators including incidence, prevalence, deaths, DALYs, years of life lost (YLL), years lived with disability (YLD), life expectancy and health-adjusted life expectancy (HALE) for the period between 1990 and 2019. Previous studies have described the methodological construction of the GBD 2019 [16]. The GBD 2019 study also categorized diseases and injuries into four-level strata, with each level consisting of mutually exclusive causes. Level 1 causes are grouped into three groups: communicable, maternal, neonatal and nutritional diseases (CMNNDs); non-communicable diseases (NCDs); and injuries. Levels 2, 3 and  [17]. The GBD has been noted to have used all available data sources and, having assessed the quality of each source to eliminate biases, used sound statistical modeling tools and methods to generate 95% uncertainty intervals (UIs) indicating the 2.5th and 97.5th percentiles in the distribution around the mean estimates [17]. Our study adhered to the guidelines for accurate and transparent health estimate reporting (GATHER) [16,17].

Country Inclusion Criteria
The GBD study does not categorize LICs, so we used the World Bank categorization, which includes 28 nations, including Afghanistan, Burkina Faso, Burundi, Central African Republic (CAR), Chad, Democratic Republic of the Congo (DRC), Eritrea, Ethiopia, Gambia, Guinea, Guinea-Bissau, Liberia, Madagascar, Malawi, Mali, Mozambique, Niger, Democratic People's Republic of Korea (North Korea), Rwanda, Sierra Leone, Somalia, South Sudan, Sudan, Syrian Arab Republic (SAR), Togo, Uganda, Yemen and Zambia. The World Bank's 2023 classification of LICs considers countries with a gross national income (GNI) per capita lesser than or equal to 1085 US dollars in 2021, and all 28 countries included in that classification are included in this study [2]. The classification of lower middle-income countries includes countries with a GNI per capita between 1086 and 4255 US dollars, upper middle-income countries as those with a GNI per capita between 4256 and 13,205 US dollars, and high-income countries as those with a GNI per capita of 13,205 US dollars or more. Further information about methods used by the World Bank to determine income classification has been discussed exhaustively in a previous article by Fantom and Serajuddin in 2016 [18].

Data Processing
This study included estimates of YLDs, YLLs and prevalence at ages 65-74. YLDs are calculated by multiplying prevalence estimates of a defined disease by a corresponding disability weight; YLLs are estimated by multiplying the number of deaths by the remaining life expectancy at the age of death. Prevalence has been noted to be aggregated in estimation at the level of individuals who may have more than one sequela or disease in this study [17]. Health-adjusted life expectancy (HALE) and life expectancy included in this study are GBD estimates of life expectancy at birth and HALE at birth for all countries. Unhealthy life years (ULYs) were calculated by subtracting HALE from life expectancy for every year (ULY = life expectancy -HALE). The proportion of ULYs on life expectancy was computed as a percentage ((life expectancy -HALE)/life expectancy).
To estimate the performance of healthcare systems in the LICs, we used the healthcare access and quality (HAQ) index, a measure that has been found to have stronger convergency validity when compared to other health-system indicators. The methods and analytic framework of the HAQ estimation have been published elsewhere [19]. The HAQ was calculated for three select age groups: young (0-14 years), working (15-64 years) and post-working (65-74 years). This study used the HAQ index of the post-working age group.
All data used in this study to estimate YLLs, YLDs, prevalence and HAQ are downloadable at the global health data exchange (http://ghdx.healthdata.org/gbd-results-tool (accessed on 2 February 2023)).

Measures
To estimate the average annual percentage change (AAPC), we utilized the joinpoint regression program version 4.9.1.0. The Joinpoint regression program is a trend analysis software developed by the US National Cancer Institute for the analysis of data from the surveillance epidemiology and results program [20,21]. The program fits a series of joined straight lines on a logarithmic scale, and segments are joined at "joinpoints". The joinpoints range from 0 to 5. The slope of each line segment of the best-fitting model was expressed as the annual percentage change (APC), and AAPC was expressed as a summary measure over a fixed interval. The joinpoint methodology for the estimation of APCs and AAPCs has been documented in previous publications [22,23].   3)), Niger (3.8 (2.9-4.6)), Eritrea (3.6 (3.4-3.9)) and Yemen (3.5 (3.4-3.6)). Similar increases in YLDs due to NCDs were recorded, with Somalia's AAPC of 4.2 (4.2-4.3) remaining the highest rate of increase, followed by Togo (3.8 (3.7-3.9)), Eritrea (3.6 (3.3-3.8)), Niger (3.5 (2.6-4.4)) and Yemen (3.3 (3.3-3.4)).
Overall, the AAPCs of NCDs were higher than those of CMNNDs in all the countries on all indicators. Suggesting a higher rate of increase in the burden of NCDs compared to that of CMNNDs.

Trend of Life Expectancy, Health-Adjusted Life Expectancy and Unhealthy Life Years
Life expectancy, HALE and ULYs increased in all countries during the study period. Figure 1 shows that although 10 countries had a life expectancy at birth below 50 years in 1990, all countries had raised their life expectancy, with only three countries having it below 60 in 2019. North Korea and Syria had the highest Life expectancies in the study period, with 67.97 and 68.79 in 1990, then 73.15 and 73.88 in 2019, respectively. The highest rate of increase was recorded in Ethiopia with an AAPC of 1.3 and Eritrea with 1.2, followed by Uganda and Rwanda, both with a 1.1 AAPC for life expectancy. HALE had a similar trend as life expectancy increased at seemingly similar rates throughout all the countries.    Life expectancy, HALE and ULYs increased in all countries during the study period. Figure 1 shows that although 10 countries had a life expectancy at birth below 50 years in 1990, all countries had raised their life expectancy, with only three countries having it below 60 in 2019. North Korea and Syria had the highest Life expectancies in the study period, with 67.97 and 68.79 in 1990, then 73.15 and 73.88 in 2019, respectively. The highest rate of increase was recorded in Ethiopia with an AAPC of 1.3 and Eritrea with 1.2, followed by Uganda and Rwanda, both with a 1.1 AAPC for life expectancy. HALE had a similar trend as life expectancy increased at seemingly similar rates throughout all the countries.  Table 3 shows that all countries also experienced increases in ULYs, although the proportion of these on life expectancy remained similar throughout the study. The proportion of ULYs on life expectancy stayed between 11% and 15% in all 30 years, indicating no change in the proportion of HALEs on life expectancy. Table 4 shows that although the HAQ index has increased in all 28 studied countries. Significant increases in Ethiopia from 18.22 (13.41-24.68) in 1990 to 23.13 (18.75-28.15) in 2019, implying a 49% increase in the index. Rwanda (43%), Afghanistan (25%) and the Syrian Arab Republic (SAR) (24%) all followed in percentage change. Although both Ethiopia and SAR reported absolute changes of 14.19 and 14.5, respectively. The highest HAQ index remains in the same countries throughout the period, with SAR, North Korea, Sudan and Yemen leading in terms of index scores.

Association between the HAQ and Health Care Needs
The increase in HAQ was found to be associated with high AAPCs for ULYs among countries (Figure 2). On the other hand, AAPCs for YLDs and prevalence for both NCDs and CMNNDs showed no correlation with the percentage change in HAQ, but AAPCs for YLLs of both NCDs and CMNNDs had a moderately positive correlation with HAQ percentage change (Figure 3).

Discussion
This study explored the results of the GBD 2019 study and highlighted the increasing life expectancy and HALE among LICs. We also investigated the trend of HAQ among the countries and how this has correlated with healthcare needs, including YLD, prevalence and YLL in these countries.
LICs are perceived to have relatively younger populations, although they are home to 9 percent of the world's population [24]. With projections of larger populations in the future, it is necessary to understand the current health issues facing elder people in these countries to plan for the future when these masses of young people turn old. Our findings emphasize the increasing number of prevalent cases, years of life lost, and years lived with disability in LICs for both CMNNDs and NCDs among the elderly. In his article, Boutayeb (2010) found that all over the world, both LICs and HICs were facing a growing "double burden" of CMNNDs and NCDs. He also found a highly increasing trend of NCDs in

Discussion
This study explored the results of the GBD 2019 study and highlighted the increasing life expectancy and HALE among LICs. We also investigated the trend of HAQ among the countries and how this has correlated with healthcare needs, including YLD, prevalence and YLL in these countries.
LICs are perceived to have relatively younger populations, although they are home to 9 percent of the world's population [24]. With projections of larger populations in the future, it is necessary to understand the current health issues facing elder people in these countries to plan for the future when these masses of young people turn old. Our findings emphasize the increasing number of prevalent cases, years of life lost, and years lived with disability in LICs for both CMNNDs and NCDs among the elderly. In his article, Boutayeb (2010) found that all over the world, both LICs and HICs were facing a growing "double burden" of CMNNDs and NCDs. He also found a highly increasing trend of NCDs in these populations, especially cardiovascular diseases (CVD) [25]. Our findings reaffirm this by showing a growing trend in both disease classifications. However, as YLDs and YLLs attributable to CMMNDs are reducing in some countries and having marginal increases in others, NCD measures, on the other hand, are increasing in all countries and at a higher rate than CMNNDs among elderly people.
Previous publications have indicated the difference between CMNNDs and NCDs, ideally confirming the fact that while most CMMNDs are treatable and curable, NCDs pose a bigger challenge since, as much as they are treatable, many are not curable [26]. A GBD risk factor study found that minimal improvements have been made in reducing exposure to behavioral risks, such as secondhand smoke, alcohol use and dietary risks, among others. Moreover, metabolic risks are increasing, including high body mass index, high fasting plasma glucose, and high systolic blood pressure in low social development index countries [16]. NCDs are harder to prevent and control by governments than CMNNDs, and with increasing age comes an increased risk of suffering these diseases, meaning that the constant exposure of older people is bound to cause an increasing burden of disease [26].
There have been increases in life expectancy since 1990 among all countries, and this correlates with a fall in YLLs as there have been successes in the minimization of the effect of acute diseases and an increase in survival rates for people past the age of 65 [24]. At the same time, there has been an increase in HALE in all countries, which has also been correlated with higher mean years of schooling, a higher total fertility rate, and achieving high levels of health-related millennium goals among LICs and lower middle-income countries (LMICs) [27]. However, as people grow older, there have been concerns about the fact that the proportion of their disability and/or dependency-free life continues shrinking, suggesting expanding morbidity, a scenario undesirable and dubbed a "failure of success" in previous studies [13,28,29]. This means that due to the positive AAPCs in all NCD-related health care needs, including prevalence, YLDs and YLLs, which are concurrent with an increasing life expectancy and HALE, it is inevitable to have increasing numbers in ULYs, which have managed to continuously rise even as these countries have succeeded in lengthening the life. The expansion of morbidity has, however, been documented to exert higher demands on healthcare systems [13,30]. Moreover, it is a notable fact that the proportion the ULYs have on life expectancy has not changed that much throughout all LICs, indicating that the proportion of life spent in ill health has continued to stay the same even with achievements related to longer and healthier lives.
There is an increase in the HAQ index of all the countries studied, with high percentage increases for Ethiopia and Rwanda, but the index is still low among the LICs, with all of them being below the 50-point mark of a 100-point index in 2019. Yet, CAR posted a 17.16 score, which is the lowest. Since the index is an indicator of healthcare quality, provision and access, these results show that elders, especially post-working adults, do not have adequate healthcare in LICs. A study by GBD collaborators in 2015 indicated that indeed low-SDI countries have low HAQ Index scores, which is a consistent finding with our study [19].
The low HAQ for elders in countries with an increasing number of unhealthy years yet with increasing life expectancy and HALE raises a need for policy and academic insights to improve healthcare systems to meet the changing demographic landscape.
To achieve an age-friendly healthcare system, improving healthcare access and quality for elders in these countries is essential. This will ensure more care and a possible reduction in morbidity.
This study has some notable strengths. These include the addition of evidence on the scarcely documented issue of the increasing burden of NCDs among elders in LICs. It also covers a 30-year trend, indicating average annual percentage changes in indicators of the burden of disease for all LICs. This study also documents trends in life expectancy and HALE and computes estimations of ULYs. Providing a basis for further studies in the direction of aging and health in LICs.
This study also has limitations, the major of which is the use of data from the GBD 2019 study, but the study's limitations have been critically documented [31]. The critics indicate a potential variance in the data quality, mostly among LICs, and our focus on older people in these countries makes it a more outstanding limitation as scarce data on the health of this population is available. HAQ variations are captured for only 1990 and 2019, yet there could be variations uncaptured during the 30 years studied.

Conclusions
As we acknowledge the success of health interventions in LICs in yielding increased life expectancy and HALE at birth, the health needs of elders are increasing, yet health systems demonstrate low scores when measured for access and quality towards these elders. LICs need to evolve as they eradicate CMNNDs among populations to be able to handle the worryingly increasing needs of NCD-laden elders in their populations. There is also a need to investigate further which components of health systems are ideal for increasing HAQ in LICs.