Comparative analysis of burden of topmost disorders requiring rehabilitation services: a systematic review and meta-analysis

Abstract

Background

Rehabilitation is an essential health service that should be available for all with health conditions affecting functioning in daily life. Initiatives for equitable distribution of scarce and limited rehabilitation resources are frequently guided by burden of diseases measures. Most studies on burden of diseases have laid greater emphasis on disability adjusted life years (DALYs) and rarely compared these burdens among conditions. This present study aims to systematically review and compare the burden of the top 10 GBD ranking of disorders.

Method

This review was pre-registered with PROSPERO (CRD42022316091). PubMed and Google Scholar were systematically searched as well as a manual search of grey literatures from 01/01/1990 to 29/02/2022. The results of this review were reported based on PRISMA guideline. Meta-analysis results were presented using forest plots and summary tables.

Results

A total of 11,367 studies were obtained from the searches, while the findings of 55 studies (17,753,434 participants) were reviewed. Majority of the studies were conducted in high-income-countries (56.1%) though, all the studies on neonatal disorders (100.0%) and congenital birth disorders (100.0%) came from the low-and-middle-income-countries. Neonatal disorders, stroke and ischaemic heart disease (IHD) topped the rank of disease burden. Overall, the burden of neurological disorders and their associated risk factors (aRFs), in terms of disease prevalence, were evaluated to be 36.75% while their mortality rate was 29.90%. Neurological conditions and aRFs accounted for 61.83% of total economic burden.

Conclusion

Neonatal disorders, stroke and IHD are the three most burdensome disorders. There is therefore need for greater focus of rehabilitation attention and resources in the coming decades.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-024-05987-x.

Introduction

The demand for rehabilitation services is rising globally, driven by an ageing population, the increasing prevalence of non-communicable diseases, advancements in acute medical care, and the availability of assistive products. Currently, one in three individuals lives with a health condition that could benefit from rehabilitation [1]. Notably, approximately 74% of years lived with disability (YLDs) are attributable to health conditions that impose functional limitations, which rehabilitation could address [2]. However, access to rehabilitation services is limited, particularly in low- and middle-income countries (LMICs), where more than 50% of the population lacks the necessary support [1].

In response to the growing demand for rehabilitation, the World Health Organization (WHO) launched the Rehabilitation 2030 initiative in 2017, in collaboration with member states and civil society. This initiative aims to address unmet rehabilitation needs worldwide and underscores the necessity of integrating rehabilitation into existing health systems. It affirms that rehabilitation is an essential health service that should be universally accessible.

Despite these initiatives, equitable distribution of healthcare resources remains challenging, and thus far are largely based on burden of diseases, which is assessed through indicators such as mortality, prevalence, quality-adjusted life years (QALYs) and disability-adjusted life years (DALYs) [3]. While the Global Burden of Diseases (GBD) studies primarily quantify disease burden in terms of DALYs, which reflect the overall impact of diseases on population health, this study expands the concept of burden to encompass not only the prevalence and mortality of these health conditions and their consequences on functioning but also the specific implications for rehabilitation services in terms of caregivers’ strain and cost.

The top Global Burden of Diseases (GBD) rankings highlight disorders with most significant global impact. However, most studies [4, 5], on the GBD have primarily focused on DALYs, incidence, prevalence, and mortality rates, often neglecting the economic burden and caregiver strain associated with disabilities. Additionally, systematic reviews [6, 7], have typically concentrated on specific disorders without comparing their burdens or exploring how these burdens relate to rehabilitation services. This study aims to systematically review and compare the burden of the top 10 GBD-ranked disorders [8], with a particular emphasis on how these burdens influence rehabilitation needs and resource allocation. By synthesizing implications for rehabilitation services, caregivers, and therapists, we will address capacity building and training needs in alignment with the Rehabilitation 2030 initiative. By highlighting these gaps in the current literature, we hope to contribute valuable insights that enhance the accessibility and effectiveness of rehabilitation services in addressing the distinct burdens associated with these conditions.

Methods

(Registration with PROSPERO—CRD42022316091)

Search strategy and database

A two-step search strategy was used to identify all relevant literature. First, electronic databases such as PubMed, and Google Scholar were systematically searched. Second, a manual search of grey literatures was conducted to identify additional relevant studies. All electronic sources were searched from January 1, 1990 to February 31, 2022. As of the publication date, no major studies appear to have been published on the subject since the end of our inclusion period. The search utilized the following terms and phrases: “burden of diseases”, “burden of disabilities”, “Global burden of diseases”, “incidence of disorders”, “prevalence of disorders”, “mortality rate of disorders”, “DALYs of disorders”, “economic burden of disorders”, and “caregivers strain of disorders”. Top 10 GBD-ranked disorders (neonatal disorders, ischaemic heart disease, stroke, lower respiratory infections, diarrhoeal diseases, COPD, road injuries, diabetes, low back pain, and congenital birth defects) [8], were used to replace diseases, disorders or disability as search terms. These disorders were selected based on a panel discussion, which concluded that all ten disorders necessitate substantial rehabilitation support (based on functional impact, clinical guidelines, and rehabilitation needs), thereby justifying their inclusion in this study. Boolean operators like “AND” and “OR” were used to combine search terms (Supplementary 1).

Eligibility criteria

Studies that were considered eligible for this review and included were determined using PECOS:

P (Populations): Patients with neonatal disorders (NDs), Ischaemic Heart Disease (IHD), stroke, Lower Respiratory Infections, diarrhoeal diseases, Chronic Obstructive Pulmonary Diseases (COPD), Road Injuries (RTI), diabetes (DM), Low Back Pain (LBP), or Congenital Birth Defects (CBDs).

E (Exposure): Studies that examined disease burden.

C (Context): Studies covering primary and referral healthcare (hospital-based).

O (Outcomes): Studies that reported incidence, prevalence, mortality rate, DALYs, Cost or caregivers’ strain.

S (Study Design): Observational Studies (Cohorts or Cross-sectional Studies).

Selection and quality assessment

Data (supplementary 2) were extracted by two authors using a pre-piloted and standardized data extraction format prepared in Microsoft Excel. For each included study, the following data were extracted: study ID (the names of the authors and year of publication), study area categorized as high-, middle-, or low-income according to the World Bank classifications—high-income countries are those with a gross national income (GNI) per capita of $12,536 or more, middle-income countries have a GNI per capita between $1046 and $12,535, and low-income countries are defined as those with a GNI per capita of $1045 or less. Additionally, information on study design, sample characteristics (size, age, sex, etc.), outcomes [incidence, prevalence, mortality, disability-adjusted life years (DALYs), cost, caregiver strain], and outcome measures used for their assessment, findings, conclusions, and methodological quality were collected. The quality of each included study was assessed using the NIH scale [9].

Ethical approval

This was sought and obtained from the Research and Ethical Committee of the University of Nigeria Teaching Hospital, Enugu.

Data analysis

To obtain and compare the meta-weighted burden of disease, the following were employed:

• i.A meta-analysis using random-effects model was employed. The random-effects model accounts for variability between studies by allowing each study to have its own effect size. It is particularly useful when the studies being analyzed are heterogeneous [10].
• ii.Cochran’s Q chi-square statistics and I² statistical test were conducted to assess the random variations between primary studies. The Cochran’s Q statistic assesses whether observed variances in study results are greater than what would be expected by chance alone; while the I² statistic quantifies the percentage of variation across studies that is attributable to heterogeneity rather than chance [11].
• iii.Publication bias was assessed objectively using the Egger bias test. This test evaluates the relationship between effect sizes and their standard errors, helping to identify whether smaller studies (often with less favourable results) are disproportionately absent from the literature. A significant result from the Egger test may indicate that publication bias could affect the overall conclusions drawn from the meta-analysis [12].
• iv.All data was analysed using MedCalc Statistical Software version 20.105 (MedCalc Software Ltd, Ostend, Belgium; https://www.medcalc.org; 2022). MedCalc is a user-friendly statistical package that offers extensive support for a variety of statistical tests, including meta-analysis. Its capabilities allowed us to efficiently conduct the necessary calculations and visualize the results.
• v.
  Post-hoc analysis was performed using mathematical operations to calculate the average burden for each of the categories (e.g. cost, caregiver’s strain, DALYs etc.) with respect to a specific condition requiring rehabilitation services (e.g. Stroke, COPD, IHD etc.). A similar operation (equation II) was also used to calculate the overall average of blocs of disorders (e.g. Neurological disorders and the associated risk factors—N + aRF, and other disorders—Ods). N + aRF = stroke, RTI (including Traumatic Brain Injuries—TBI and Spinal Cord Injuries—SCI), IHD, and Diabetes Mellitus (DM).

  $$\text{a}=\sum \left(\text{v}\ast \text{n}\right)/\sum \text{n}$$ 1

where a is the Average burden for a particular disorder (e.g. DALYs of DM). v is the Value of the burden for a particular disorder obtained from a study (e.g. DALYs of DM from Tandon et al. [84]). n is the Total number of participants in the study (e.g. [84]). ∑n is the Total number of participants in all the studies that assessed the burden for the particular disorder (eg for DALYs of DM—the summation of n from Tandon et al. [84], Li et al. [85], Oliveira et al. [86] etc.)

$${\text{A}}^{\text{b}}=\sum \left({\text{a}}^{\text{b}}\ast \sum {\text{n}}^{\text{b}}\right)/\sum {\text{N}}^{\text{b}}$$ 2

where A^b is the Average burden (e.g. DALYs) for a bloc of disorders (eg N + aRF). a^b is the Average burden for each disorder in the bloc (e.g. DALYs of Stroke, IHD, DM etc. for N + aRF). ∑n^b is the Total number of participants in all the studies that assessed the burden for each of the disorders in the bloc (e.g. for DALYs of DM—the summation of n from Tandom et al. [84], Li et al. [85], Oliveira et al. [86] etc.). ∑N^b is the Summation of total participants for each disorder in the bloc.

Data synthesis and reporting

The results of this review were reported based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline (Table 1). Also, Meta-analysis results were presented using forest plots and summary tables.

Table 1.

PRISMA checklist for systematic review and meta-analysis of rehabilitation service burden

Item	Description	Location in Manuscript
1	Title	Page 1, Line 1
2	Abstract: Structured abstract including background, methods, results, and conclusions	Page 1, Lines 3–22
3	Introduction: Rationale for the review and objectives clearly stated	Page 2, Lines 1–10
4	Methods: Protocol and registration (PROSPERO number)	Page 3, Line 1
5	Eligibility criteria: Clearly defined (PECOS)	Page 4, Lines 1–4
6	Information sources: Databases searched and time frame noted	Page 4, Line 5
7	Search strategy: Detailed search terms and Boolean operators used	Page 4, Lines 6–10
8	Selection process: Flowchart detailing study selection (PRISMA flowchart mentioned)	Page 6, Line 1
9	Data extraction process: Standardized data extraction format described	Page 5, Lines 10–12
10	Quality assessment: Methodological quality assessed using the NIH scale	Page 5, Line 13
11	Results: Study selection: Number of studies included and excluded, with reasons	Page 6, Lines 2–5
12	Study characteristics: Data on included studies (sample size, demographics)	Page 7, Lines 1–5
13	Synthesis of results: Meta-analysis results with pooled prevalence and mortality rates	Page 8, Lines 1–10
14	Risk of bias: Assessment results mentioned	Page 7, Line 12
15	Discussion: Summary of main findings	Page 9, Lines 1–5
16	Implications for practice and research discussed	Page 9, Lines 6–10
17	Limitations of the review discussed	Page 10, Lines 1–5
18	Conclusions drawn from the findings	Page 10, Lines 6–10
19	Funding: Source of funding for the review stated	Page 11, Line 1
20	Conflicts of interest: Disclosure of any potential conflicts of interest	Page 11, Line 2

Results

A total of 11,419 studies, were obtained from the electronic searches of the databases, while the findings of 107 studies with available full text articles were synthesized and presented [13–119]. Details are presented in the PRISMA flowchart (Fig. 1). A self-assessment of the article’s quality using the AMSTAR-2 scale is presented in Table 2.

Fig. 1.

PRISMA flow diagram for study selection in the systematic review of rehabilitation service burden

Table 2.

AMSTAR-2 checklist for assessing methodological quality in systematic reviews of rehabilitation service burden

AMSTAR-2 criteria	Assessment	Comments
1. Protocol registration	Yes	Review was pre-registered with PROSPERO (CRD42022316091)
2. Study selection	Yes	Clear eligibility criteria based on PECOS (Participants, Exposure, Context, Outcomes, Study Design)
3. Comprehensive literature search	Yes	Systematic search of multiple databases and manual search for grey literature conducted
4. Study characteristics	Yes	Characteristics of included studies reported (sample size, study design, outcomes)
5. Risk of bias assessment	Yes	Methodological quality assessed using the NIH scale; with detailed risk of bias assessment
6. Data synthesis	Yes	Meta-analysis conducted and results summarized using forest plots and tables
7. Funding and conflicts of interest	Yes	Explicit statement regarding no funding sources and potential conflicts of interest
8. Quality of evidence	Yes	Overall quality of evidence discussed, highlighting limitations
9. Discussion of limitations	Yes	Limitations of the review addressed, including study distribution concerns
10. Conclusion supported by results	Yes	Conclusions are supported by the data presented in the results section

Methodological qualities of the included studies

In general, the majority of the studies included in this review (57.0%) were moderate quality trials (NIH Scores 6 to 9). Majority of the diarrhoea (85.7%), stroke (72.2%), and ischaemic heart disease (69.2%) based studies had moderate quality, though most of the studies on lower respiratory (60.0%), congenital birth defects (60.0%), neonatal disorders (57.1%) and COPD (57.1%) had a high methodological quality as shown in Table 3.

Table 3.

Overview of Methodological Quality Ratings of Included Studies by Condition Assessed (N = 55)

Conditions	High quality (≥ 10)			Moderate quality (6–9)			Low quality (≤ 5)			Total
	Studies	n	%	Studies	n	%	Studies	n	%
Neonatal disorders	[62, 100, 101, 103]	4	57.1	[68, 93, 94]	3	42.9	–	0	0	7
Ischaemic heart diseases	[17, 20, 22, 30]	4	30.8	[38, 40, 43, 56, 69–72]	9	69.2	[34]	1	8.3	13
Stroke	[24, 28, 42, 74, 106]	5	27.8	[31, 33, 39, 46, 60, 73, 92, 97, 104, 105, 114–116]	13	72.2	–	0	0	18
Lower respiratory disorders	[15, 30, 108]	3	60.0	[102, 107]	2	40.0	–	0	0	5
Diarrhoea	[66]	1	14.3	[63–67, 75, 76]	6	85.7	–	0	0	7
COPD	[14, 47, 50, 51, 53, 57, 58, 78]	8	57.1	[25, 41, 49, 77, 109, 110]	6	42.9	–	0	0	14
Road injuries	[79, 80, 98, 99, 111]	5	50.0	[81–83, 95, 96]	5	50.0	–	0	0	10
Diabetes mellitus	[13, 16, 27, 29, 35, 36, 52, 112, 113]	9	28.1	[18, 19, 37, 44, 45, 48, 55, 61, 84–88]	13	56.5	[32]	1	4.3	23
Low back pain	[90]	1	20.0	[59, 89, 118]	3	60.0	[117]	1	20.0	5
Congenital Birth defects	[21, 26, 119]	3	60.0	[54, 91]	2	40.0	–	0	0	5
Total		43	40.2		61	57.0		3	2.8	107

Study distribution based on location and economy

Majority of the studies involved in this review (58.0%) were conducted in high income countries, whereas very few studies (6.7%) were carried out in low- income countries as shown in Table 4. Majority of the studies that bordered on road injuries (90.0%), lower respiratory disorders (77.8%), COPD (71.4%), stroke (66.6%), diabetes (65.3%), low back pain (60.0%) were performed in high income countries. On the contrary, most of the studies on neonatal disorders (84.6%) and congenital birth disorders (80.0%) came from the low- and middle- income countries (LMICs) as shown in Fig. 2.

Table 4.

Analysis of study locations and economic contexts in the reviewed literature (N = 55)

Conditions		HIC						uMIC						lMIC					LIC
		Studies/Refs			n	%		Studies/Refs			n		%	Studies/Refs		n	%		Studies/Refs		n	%
Neonatal Disorders		[68, 94]			2	15.4		[62, 93, 101]			3		23.1	[62, 63, 100, 103]		4	30.8		[62(4)]		4	30.8
Ischaemic heart Diseases		[23, 34] [40(5)] [43, 56]			9	50.0		[17, 20, 38, 69–72]			7		38.9	[22]		1	5.6		[40]		1	5.6
Stroke		[31, 33, 39, 46, 60, 74, 92, 97, 104–106, 114]			12	66.6		[28, 73, 115]			3		16.7	[24, 42, 116]		3	16,7		–		0	0
Lower Respiratory Disorders	[102(5)] [107, 108]		7	77.8			[15]		1	11.1		[30]			1			11.1		–	0	0
Diarrhoea	[63, 65, 67]		3	42.9			[64]		1	14.3		[75, 76]			2			28.6		[66]	1	14.3
COPD	[14, 21, 41, 47, 50, 51, 53, 57, 58, 109]		10	71.4			[78, 110]		2	14.3		[49, 77]			2			14.3		–	0	0
Road injuries	[80–83, 95, 96, 98, 99, 111]		9	90.0			–		0	0		[79]			1			10.0		–	0	0
Diabetes Mellitus	[13, 16, 19, 27, 32, 35–37, 44, 45, 48, 55, 88, 112, 113]		15	65.3			[18, 29, 52, 61, 85–87]		7	30 4		–			0			0		[84]	1	4.3
Low Back Pain	[59, 89, 117]		3	60.0			[90]		1	20.0		[118]			1			20.0		-	0	0
Congenital Birth Defects	[91]		1	20.0			[26, 119]		2	40.0		[21]			1			20.0		[54]	1	20.0
Total			69	58.0					26	21.8					16			13.4			8	6.7

COPD chronic obstructive pulmonary disease, LIC low income country, lMIC lower middle income country, uMIC upper middle income country, HIC high income country

Fig. 2.

Stacked bar chart illustrating the distribution of studies by economic classification according to world bank regions (COPD chronic obstructive pulmonary disease, LIC low income country, lMIC lower middle income country, uMIC upper middle income country, HIC high income country)

Data source, risk of bias and outcomes assessed in the included studies

Majority of the studies included in this review sourced their data directly (primary source) from the participants (50.5%) while 46.7% of the studies relied on secondary data from institutional data banks. Also, most of the studies (57.0%) had a moderate risk of bias and the majority of them assessed disease prevalence (27.5%), the mortality rate (25.4%), and disability-adjusted life years (18.1%) as shown in Table 5.

Table 5.

Overview of data sources, risk of bias, and outcomes evaluated in included studies

Variable	Categories	f	%
Data source	Primary	54	50.5
	Secondary	50	46.7
	Both	3	2.8
Study population	Regional	13	12.2
	National	89	83.1
	International	3	2.8
	Continental	2	1.9
Risk of bias	Low	43	40.2
	Moderate	61	57.0
	High	3	2.8
Outcomes assessed	Prevalence	38	27.5
	Incidence	7	5.1
	Mortality	35	25.4
	Cost	21	15.2
	Care Giver Strain/Burden	12	8.6
	DALYs	25	18.1

Meta-synthesis of results

Pooled prevalence of disorders

Seven disorders (diabetes, IHD, stroke, COPD, diarrhoea, low back pain and road injuries) met the criteria for meta-analysis. A total of 1.3 billion persons were involved in the meta-analysis for diabetes, the pooled prevalence was 10.31% as shown in Fig. 3. For ischaemic heart diseases and stroke, a total of 108,688 and 1.07 million participants were involved in their meta-analyses, and the pooled prevalence were 2.33% and 4.44% as shown in Figs. 4 and 5 respectively. Meta-analysis for COPD and diarrhoea revealed that a total of 181,488 and 9811 participants respectively were involved and the respective pooled prevalence were 7.04% and 9.08%as shown in Figs. 6 and 7 respectively. Finally, a total of 377,521 and 1.1 million participants were involved in the meta-analysis for low back pain and road injuries, the pooled prevalences were 41.00% and 10.1% respectively as shown in Figs. 8 and 9.

Fig. 3.

Forest plot representing the meta-analytic pooled prevalence of diabetes

Fig. 4.

Forest plot representing the meta-analytic pooled prevalence of ischaemic heart diseases

Fig. 5.

Forest plot representing the meta-analytic pooled prevalence of stroke

Fig. 6.

Forest plot representing the meta-analytic pooled prevalence of COPD

Fig. 7.

Forest plot representing the meta-analytic pooled prevalence of diarrhoea

Fig. 8.

Forest plot representing the meta-analytic pooled prevalence of low back pain

Fig. 9.

Forest plot representing the meta-analytic pooled prevalence of road injuries

Pooled mortality rates of disorders

Only six disorders (diabetes, IHD, stroke, diarrhoea, congenital birth defects and neonatal disorders) met the criteria for meta-analysis. A total of 403,806 persons were included in the meta-analysis for diabetes; their pooled mortality rate was 1.79% as shown in Fig. 10. For ischaemic heart diseases and stroke (IHD vs. Stroke), the total participants were 662,851 vs. 687,025 and their pooled mortality rates were 4.31% vs. 1.34% as shown respectively in Figs. 11 and 12. Finally, for diarrhoea, congenital birth defects and neonatal disorders (Diarrhoea vs. CBD vs. NDs), the total participants were 85,214 vs. 5669 vs. 246,320 while their pooled mortality rates were 10.00% vs. 14.61% vs. 4.09% respectively as shown in Figs. 13, 14 and 15.

Fig. 10.

Forest plot representing the meta-analytic pooled mortality rate of diabetes

Fig. 11.

Forest plot representing the meta-analytic pooled mortality rate of ischaemic heart diseases

Fig. 12.

Forest plot representing the meta-analytic pooled mortality rate of stroke

Fig. 13.

Forest plot representing the meta-analytic pooled mortality rate of diarrhoea

Fig. 14.

Forest plot representing the meta-analytic pooled mortality rate of congenital birth defects

Fig. 15.

Forest plot representing the meta-analytic pooled mortality rate of neonatal disorders

Average cost burden of disorders

Diabetes ($33,153.95) and stroke ($29,424.74) had the highest annual cost burden per patient while the lowest were lower respiratory disorders ($1495.00) and ischaemic heart diseases ($3824.00) as shown in Table 6.

Table 6.

Overview of cost burden associated with specific health conditions

Conditions	Studies	n	Cost ($)
			Value	Average
Diabetes Mellitus	Rapattoni et al. [16]	811,794	34,913.50
	Talamantes et al. [19]	15,000	4563.90
	Guo, et al. [29]	2853	865.00
	Almutairi et al. [44]	2300	2175.05
	Sicras-Mainar et al. [45]	26,845	4903.80
	Li et al. [61]	871	2127.10
	Holmes et al. [113]	1578	981.97
	Total	861,241		33,153.95
IHD	Kumar et al. [22]	100	3824.00	3824.00
Stroke	Abdo et al. [28]	203	6965.00
	Zhao et al. [33]	2158	31,537.50
	Total	2361		29,424.74
COPD	Lisspers et al. [14]	17,479	14,496.90
	Verberk et al. [25]	213	7272.38
	Wacker et al. [41]	2139	7989.30
	Sicras-Mainar et al. [51]	930	4030.87
	Merino et al. [57]	386	3384.70
	Mulpuru et al. [58]	1894	1280.00
	Tachkov et al. [78]	426	14,348.40
	Total	23,467		12,171.19
Diarrhoea	Magee et al. [61]	83,882	27,408.00
	Townsend et al. [1]	143,152,000	949.82
	Total	143,235,882		965.31
LRD	Thanaviratananich et al. [15]	1000	1495.00	1495.00
Road injuries	Collie et al. [81]	720,800	14,705.88
	Scholten et al. [82]	366,240	857.30
	Total	1,087,040		10,040.11

IHD Ischaemic heart diseases, LRD lower respiratory disorders

Average = ∑(value * n)/n

Summary of caregivers burden, DALYs and incidence of disorders

Road injuries mainly SCI and TBI (64.18%) and COPD (49.32%) had the highest pooled burden on caregivers as shown in Table 7. On the other hand, stroke (1826.48/100,000) and road injuries (1763.50/100,000) were reported to have the highest DALYs as shown in Table 8. Going by the few reports on incidence, stroke (7.00%) and diarrhoea (5.20%) had the highest pooled incidence rates as shown in Table 9.

Table 7.

Overview of caregivers’ strain associated with specific health conditions

Conditions	Studies	n	Burden of Care (%)
			Value	Average
Stroke	Chuluunbaatar et al. [24]	203	23.35
	Choi et al. [46]	78	19.89
	Bugge et al. [104]	110	40.70
	Isle et al. [105]	90	29.97
	Oosteveer et al. [106]	179	22.91
	Total	660		20.73
LRD	McPeake et al. [107]	47	44.68
	Neri et al. [108]	225	15.11
	Total	272		20.22
COPD	Marcus et al. [109]	90	51.33
	Pinto et al. [110]	42	45.00
	Total	132		49.32
Road injuries	Schloten, et al. [111]	67	64.18	64.18
Diabetes Mellitus	King et al. [112]	795	18.40	18.40

Average = ∑(value * n)/n

Table 8.

Overview of disability-adjusted life years (DALYs) for specific health conditions

Conditions	Studies	Country	Population (n)	DALYs (/100,000)
				Value (v)	Average (a)
Neonatal Disorders	Kim et al. [68]	Korea	51,220,000	64	64
IHD	Zhang et al. [69]	China	1,386,000,000	1760.2
	Zhang et al. [70]	China	1,393,000,000	114.8
	Yu et al. [71]	China	1,402,000,000	313.9
	Xie et al. [72]	Fujia, China	3,385,567	1202.7
	Total		4,184,385,567		727.40
Stroke	Sudharsanan et al. [42]	India	1,366,000,000	2984
	Yu et al. [71]	China	1,402,000,000	791.9
	Gao et al. [73]	China	1,393,000,000	1762
	Chen et al. [74]	Taiwan	23,374,000	77.2
	Total		4,184,374,000		1826.48
Diarrhoea	Townsend et al. [75]	India & China	2,667,000,000	1787.1
	Ogbo et al. [76]	Nigeria	201,000,000	612.9
	Total		2,868,000,000		1704.81
COPD	Adhikari et al. [77]	Nepal	28,100,000	2143.5
	Tachkov et al. [78]	Bulgaria	7,076,000	5.2
	Total		35,176,000		1713.36
Road Injury	Rahimi-Movaghar et al. [79]	Iran	73,760,000	70.0
	Hall et al. [80]	USA	328,300,000	8.7
	Collie et al. [81]	Australia	22,030,000	28,822.3
	Scholten et al. [82]	Netherlands	16,870,000	7.1
	Te-Ao et al. [83]	New Zealand	4,609,000	3852.0
	Total		371,809,000		1763.50
Diabetes Mellitus	Tandon et al. [84]	India	1,353,000,000	792
	Li et al. [85]	China	1,357,000,000	1912
	Oliveira et al. [86]	Brazil	193,900,000	78.7
	González et al. [87]	Argentina	43,130,000	57.6
	Bener et al. [88]	Qatar	2,459,000	4.35
	Total		2,949,489,000		1249.00
LBP	Cuschierie, et al. [89]	Malta	525,285	716
	Wu et al. [90]	China	1,412,600,000	411
	Total		1,413,125,285		411.11

Average = ∑(value * n)/n

Table 9.

Analysis of incidence rates associated with selected health conditions

Conditions	Studies	n	Incidence (%)
			Value	Average
Diarrhoea	Thibault et al. [67]	278	5.20	5.20
Road injuries	Scholten et al. [82]	34,681	0.2136
	Burke et al. [98]	14,619	0.00271
	Noonan et al. [99]	1,785	0.0053
	Total	51,085		0.15
CBD	Kortbeek et al. [91]	10,800	0.18	0.18
Stroke	Koton et al. [97]	14,357	7	7.00

Average = ∑(value * n)/n

Comparison of burden of neurological disorders with other disorders

The burden of neurological disorders and their associated risk factors relative to other conditions in terms of disease prevalence were evaluated to bear 23.60% of this burden while their mortality rate bore 29.81% of the burden. Further evaluation revealed that neurological conditions and associated risk factors accounted for 79.67% and 49.76% of total cost and caregivers burden respectively. Relative to other conditions, neurological disorders were responsible for 52.16% and 51.10% DALYs and the incidence burden of disorders respectively as shown in Fig. 16. Summarily, neurological disorders accounted for 48.68% of the overall burden of disorders.

Fig. 16.

Burden comparison of neurological disorders and associated risk factors (RFs) versus other disorders

Ranking of various conditions based on burden of disease

Judging by the GBD ranking, disease prevalence, mortality, cost, caregivers strain DALYs and incidence as synthesized from this study, neonatal disorders remained top-ranked. However, stroke and road injuries (mostly SCI and TBI) surged higher for the second and third positions respectively. The mid positions were occupied by diarrhoea (fourth), lower respiratory disorders (fifth), and diabetes (sixth). The least ranked disorder based on the totality of the burden of diseases was congenital birth disorders as shown in Fig. 17.

Fig. 17.

Hierarchical comparison of disorders based on their disease burden

Discussion

Rehabilitation is inarguably an essential component of universal health coverage, focusing on achieving functional independence in daily activities and societal participation, which are fundamental for a healthy and functional global population [120]. The top ten global burden of diseases (GBD) studied in this review are capable of limiting functional activities and restricting participation. Therefore, rehabilitation services are invaluable in alleviating the burden of disability.

For those aspects, that could be meta-analytically analysed, i.e. prevalence and mortality rates, the analyses revealed a high degree of heterogeneity among the included studies. Several factors may contribute to this variability. First, differences in study populations, such as catchment population, age, socioeconomic status, and baseline health conditions, can significantly influence these measures.

Additionally, the low proportion of included studies compared to the total initial studies may reflect the stringent inclusion criteria based on the WHO criteria for GBD. While this ensures the quality and relevance of the studies included, it also limits the number of available studies for analysis. Notably, the limited high-quality publications on conditions such as diarrhoea and low back pain may suggest these conditions are less prioritized in current research agendas [121], or it is possible that existing studies did not meet our eligibility criteria.

Despite the high heterogeneity, the inclusion of several high-quality studies in our review [13–17, 20–30, 34–36, 50–53, 78–80, 98–101, 111–113, 119], provides a solid foundation for policy-making regarding rehabilitation for the selected conditions, particularly neurological disorders. This underscores the need for further research to standardize methodologies and broaden the understanding of rehabilitation’s role in managing the global burden of disease. Addressing these discrepancies in future studies will be crucial for enhancing the effectiveness and applicability of rehabilitation interventions across diverse populations and conditions.

The distribution of the included studies, based on location, revealed the unequal distribution of quality research output in favour of the high- income economies [122]. There remains a wide gulf between quality research publications emanating from the high- income countries and the low- and middle- income countries. Another critical observation is the type of studies coming from the different economic belts. The studies from LMICs were mostly paediatric conditions such as neonatal disorders and congenital birth disorders [21, 26, 54, 62, 63], which could be a reflection of the low level pre- and peri-natal healthcare programmes in this region [123, 124]. On the contrary, the studies that emanated from the high- income countries were those mostly acquired in adulthood, which could be secondary to the lifestyle patterns in such economies as well as enhanced longevity. The implication of this disparity to the LMICs could be a higher cost of healthcare, a longer burden of care as well as greater need for rehabilitation services for conditions that may require a lifetime care.

In evaluating the disease burden across different economies, it is essential to recognize the potential biases that can influence assessments between high-income countries (HICs) and low- and middle-income countries (LMICs). HICs typically benefit from advanced healthcare infrastructure, better access to services, and more comprehensive health data collection systems, leading to more accurate diagnosis and reporting of disease burdens. In contrast, LMICs often face challenges such as underreporting, incomplete data, and limited access to healthcare resources, which can distort the true burden of diseases prevalent in these regions. Moreover, socioeconomic factors play a significant role; LMICs may experience higher rates of infectious diseases and maternal-child health issues, while HICs are more likely to report chronic conditions such as cardiovascular diseases. Cultural perceptions and stigmas surrounding certain health issues can further complicate reporting, particularly in LMICs where mental health conditions may be underrecognized. Additionally, research funding often prioritizes diseases prevalent in HICs, leading to a lack of focus on conditions that disproportionately affect LMIC populations. These disparities underscore the need for a more nuanced understanding of disease burden that considers the unique challenges and resources of different economies, ensuring that health policies and resource allocations are effectively tailored to meet the needs of diverse populations.

There appear to be no known reported advantage of results from data obtained directly from the patients and those obtained from secondary data. However, data obtained in the context of a primary study may be more structured along with the objectives and purposes of the study, and less exposed to misinterpretation, compared to secondary data. The low risk of bias among the included studies may strengthen the outcome of this review. Although this review focused on the burden of disorders, assessing mortality rate and disease prevalence were the two highest foci in most of the included studies. Moreover, understanding patient outcomes before and after rehabilitation is crucial for contextualizing these findings.

While the burden of disease is significant, rehabilitation can lead to substantial improvements in functional independence, quality of life, and physical and cognitive function as well as emotional well-being [125]. The available evidence highlights the importance of integrating rehabilitation into the management of these disorders, as effective interventions can significantly enhance patients' daily functioning and overall well-being. Recognizing these dynamics not only enriches our understanding of the burden but also underscores the potential benefits of targeted rehabilitation efforts in improving health outcomes for affected populations.

Low back pain had the highest pooled prevalence among the seven disorders involved in the meta-analysis. This should raise more concern as it is most common among the workforce, including the health care workers who attend to them [126]. The negative effects on productivity and the exponential adverse health risks from prolonged and irrational use of painkillers have compounded this disorder [127, 128]. In Africa, the rising prevalence of low back pain is almost at par with the developed nations [129]. It is high time we recognized the early phase of musculoskeletal disorders which could herald a major bone degenerative disease, the complications of which are debilitating, such as paraplegia, urinary incontinence and their attendant risk of pressure ulcers and deep vein thrombosis or the fatal pulmonary embolism [130]. These may be difficult to manage in the LMICs where health insurance is still in its infancy and patients have to pay out of pocket for their health care needs [131]. Education on primary prevention and ergonomics should be emphasized alongside rehabilitation. Diabetes was the disorder with the next highest burden based on prevalence. The high prevalence of diabetes would mean that non-communicable diseases and unhealthy lifestyles are also persistent sources of burden to people and mitigating measures should be put in place to curb them. The implication for rehabilitation is that most of the available resources, time and personnel will be directed towards rehabilitating the conditions as well as the consequences of the disorders.

The mortality rate of congenital birth defects is markedly high when compared with the other six disorders involved in the meta-analysis. Explanations for the high mortality may be related to aetiologies that border on infections, genetic, nutritional, and toxic causes that are most prevalent especially in low- and middle-income countries where the healthcare and more specifically, rehabilitation services are scarcely available [132]. Also, ischaemic heart disease and neonatal disorders recorded high levels of mortality rate. Efforts in rehabilitation should therefore be directed towards controlling the risk and predisposing factors of the two disorders. Although stroke and diabetes had a relatively low pooled mortality rate, it is imperative to pay close attention to them as disorders that are capable of compounding the global disease burden. This may increase the burden of care both on the professionals and the caregivers of such individuals with diabetes and/or stroke.

The cost burden of care for a disorder could be a reflection of the amount of expertise and personnel that would be required to deliver the care. Diabetes stood out as a disorder with the greatest cost burden. The direct costs of rehabilitation and other indirect costs, including those borne by caregivers, place a major burden on the individuals. Stroke had the second highest cost burden. When the cost of care becomes a significant burden on the patient, family and the community, patients may tend to seek alternative care [133].

Road injuries (SCI and TBI) gave the highest burden on caregivers. Most of these patients have prolonged hospital stays, and in particular, after discharge, lack independency and are requiring assistance at home for both physical, mental, and behavioural sequelae of their injury. It requires concerted efforts of the caregivers at the expense of their own health and risk of dismissal in their places of work because of the time they have to spend at home, taking care of their disabled relatives [134]. In addition, these spouses and children have to be involved in the rehabilitative measures to ensure effective therapy [135]. Therefore, to reduce this burden, the principle of task shifting and task sharing has to be explored to recruit more manpower into the rehabilitation procedure.

The majority of the economic losses can be attributed to stroke, and a combination of traumatic brain and spinal cord injuries from road injuries; and these losses have been projected to disproportionately affect low- and lower-middle-income countries, risking up to a 0.6% and 0.54% loss of GDP, respectively, by 2030 [136]. The disability-adjusted life years (DALYs) from neurological conditions is expected to increase in these economically deprived regions of the world.

Finally, this review showed that neurological conditions alone had almost half of the overall burden of the top 10 GBD conditions studied. These disorders are recognized leading cause of disability and second leading cause of death worldwide [137]. Emphasis is now on brain health, focusing on the prevailing risk factors at the various levels of prevention [138]. Neurorehabilitation is an aspect of rehabilitation therapy that requires skillful interventions in terms of cognitive stimulation therapy, cognitive behavioural therapy, speech and language therapy, re-training motor functions, and other neurorehabilitation procedures that require special equipment in electromagnetic stimulation therapy and motor learning [139]. The enormity of this burden and the required resources needed for rehabilitation call for pragmatic and accelerated actions by the stakeholders to train more workers who will be involved in patient enlightenment and therapeutic interventions as well as healthcare structures for neurorehabilitation, both centre- and community-based.

Therefore, it is pertinent to stress that in order to deliver quality rehabilitation services and achieve the Rehabilitation Vision 2030, capacity building directed at all cadres of health providers should be done on regular basis. A secondary preventive approach should be included as an adjunct intervention in the protocols to militate against the cardiovascular risk factors. Rehabilitation experts need to engage caregivers in order to attain continued prompt delivery of care.

Strengths, limitations and future directions

This review covered three decades which are substantial enough to be representative of the actual burden of the diseases under study. Also, the comprehensive meta-analysis for adequate comparison among the diseases and regions of the world would help to identify the gaps and areas in need of accelerated and intensive governmental and non-governmental inputs. The fewer studies from the LMICs, not giving a wide coverage, might not be a replica of the burden. Therefore, our conclusions might be too early to draw. We implore the WHO, and other non-governmental organisations to include, emphasize and implement rehabilitation activities in their programmes in order to address the teaming exponential rise in these chronic and debilitating disorders, most especially in the LMICs where resources are meager.

Conclusion

We systematically reviewed and compared the burden of the top 10 GBD-ranked disorders requiring rehabilitation. Neurological disorders along with their risk factors accounted for about one-third of disease prevalence and mortality as well as about two-thirds of the economic burden of diseases. Furthermore, neonatal disorders, stroke and road injuries (SCI and TBI) are the top three most burdensome disorders. We also identified gaps in epidemiological data particularly in low- and middle- income countries. These have implications for the provision of rehabilitation services, capacity building and training of therapists, and other personnel and resource allocation to achieve the goals of the WHO Rehabilitation 2030 initiative.

Author contributions

Conceptualization: MOO, WF, RKD, TP, & ENDE; Methodology: ENDE, MOO, & TP; Data Collection: ENDE, OUC, DTR, TBI, ABO, & SCB; Data Analysis: ENDE, OUC, BCA. Drafting the Manuscript: ENDE, BCA, OUC, PO, OOD; Critical Review: ENDE, MOO, & OUC; Supervision: MOO, WF, RKD, TP, & ENDE; Project Administration: ENDE & OUC; Final Approval: ENDE, OUC, DTR, TBI, ABO, SCB, BCA, PO, OOD, MOO, WF, RKD, & TP.

Funding

TP received a non-restricted personal grant by the BDH Bundesverband Rehabilitation e.V. (charity for neuro-disabilities); the sponsors had no role in the decision to publish or any content of the publication.

Availability of data and materials

All data generated or analysed during this study are included in this published article [and its supplementary information files].

Declarations

Ethics approval and consent to participate

Ethical approval was sought and obtained from the Research and Ethics Committee of the University of Nigeria Teaching Hospital. As this study is a secondary research (systematic review), informed consent was not warranted.

Consent for publication

The authors give their consent for the publication of this study.

Competing interests

None.