Introduction

Information concerning the burden of acute musculoskeletal conditions, coupled with experiential evidence, suggests that sequelae of injuries and infections are responsible for significant disability in low- and middle-income countries (LMICs) [1–9]. Injuries alone cause more than 5 million deaths per year [1], more than HIV/AIDS, malaria, and tuberculosis combined [2], and for every death, there may be twenty non-fatal outcomes and one case of permanent disability [3, 4], many related to the musculoskeletal system.

Alongside this growing burden comes the realization that there are gross disparities in access to surgical care globally. Weiser et al. estimated that only 3.5 % of the world’s surgical procedures are performed in countries at or below the lowest third for per capita health expenditure [10]. Inadequate access to health services has lead to increased mortality for many conditions, or to delayed or “neglected” presentations for non-fatal conditions. Such cases require treatment strategies that are more complex, more costly, and are less likely to achieve a suitable outcome. Within the realm of musculoskeletal surgery, prompt reduction of fractures and joint dislocations often eliminates the need for complex open surgical procedures. Simple drainage of an abscess and debridement of devitalized bone complicating acute osteomyelitis may reduce the risk of chronic osteomyelitis, which would require multiple surgical procedures to eradicate the infective focus, reconstruct osseous defects, and treat coexisting problems such as angular deformity and/or limb length discrepancy.

The provision of safe and timely musculoskeletal surgical services may be viewed as “primary prevention” of disability, especially at the primary referral level in LMICs where a significant percentage of the population receives their health care. While a number of previous investigations have documented deficiencies in the availability of surgical services in LMICs [11–33], none have focused on musculoskeletal care. The goal of this study is to describe the availability of musculoskeletal surgical services at different tiers of the health system in a convenience sample of 883 health facilities from 24 low- and lower-middle-income countries.

Method

A tool for situational analysis of the availability of surgical and anesthetic services at individual health facilities was developed by members of the World Health Organization’s (WHO) Global initiative for Emergency and Essential Surgical Care (GIEESC) in 2007, focusing on (1) infrastructure, (2) human resources, (3) interventions, and (4) equipment and supplies [34]. The items listed in the questionnaire have been extracted from the teaching materials from the WHO’s Emergency and Essential Surgical Care (EESC) project, and “Essential” might also be defined as those services which should be available within the context of universal access [35, 36]. Recognizing that the specific interventions will be refined by local contextual variables, these target “high priority” conditions (1) which have a large public health burden, (2) for which there is an intervention which is highly successful, and (3) for which the intervention is cost-effective and can be promoted globally [37]. The musculoskeletal interventions included in the tool are wound debridement, irrigation and drainage of abscesses, closed treatment of fractures, open treatment of fractures, joint dislocation treatment drainage of osteomyelitis/septic arthritis, amputation, and clubfoot. Selected equipment and supplies related to musculoskeletal services and included in the questionnaire include tourniquet and splints.

Since that time the tool has been utilized in more than 50 countries; facilities have been selected by the Ministries of Health, and questionnaires have been administered by representatives from the Ministries of Health, the WHO country offices, and/or GIEESC members. The forms are then sent to WHO headquarters in Geneva and entered into a database.

This database was accessed in early December 2013, at which time information was available for 1,076 health facilities in 56 countries. We chose to remove countries from which less than 10 facilities were sampled, and those facilities with incomplete datasets. We also removed data from the only high-income country (Trinidad and Tobago). The data on clubfoot were not included in the present study, having been presented in another publication.

Recognizing the variability in terminology used to describe a particular level of facility within a country’s health system, we have elected to present the data based on both the number of beds and the type of health facility, recognizing that there is some degree of overlap. Facilities have been divided into three categories based on the number of beds, according to WHO’s textbook Surgical Care at the District Hospital [38], as (1) <100 beds, (2) 100–300 beds, and (3) >300 beds. Typically, facilities with less than 100 beds are rural hospitals or health centers with minor surgical capacity, while those with more than 300 beds are tertiary level facilities that would be expected to perform more complex surgical procedures. Three levels of facility have been selected, namely (1) primary referral level (health center, district/rural/community hospital), (2) secondary/tertiary level (provincial or general hospital), and (3) Private/NGO/Mission.

With the goal of evaluating any differences between the availability of services at government facilities versus Private/NGO/Mission facilities, we performed a statistical comparison between these based on number of beds (≤100, 100–300, and >300) using a Chi-Square analysis, with statistical significance at p < 0.05.

Results

Our final dataset included 883 health facilities from 15 Low- and 9 lower-middle income countries, representing one fourth of the world’s population (Table 1). Figure 1 illustrates the levels of facility for each level of beds. A subset of our data have been included in previous studies utilizing the GIEESC tool.

Table 1 Selected indicators
Full size table
Fig. 1
figure 1

Type of health facility and number of beds. We chose to analyze out data based on (1) type of health facility as Primary (health center, district/rural/community hospital) (Black), Secondary/Tertiary (provincial or general hospital) (Gray), and Private/NGO/mission, and (2) the number of beds (≤100, 100–300, and >300) (white). This figure illustrates the degree of overlap between the groups in our analysis

Full size image

Data concerning the availability of infrastructure, supplies, and human resources for the entire group, type of facility, and number of beds are shown in Table 2. Deficiencies were most pronounced in facilities with ≤100 beds or primary referral level facilities, but were also identified at higher levels of service delivery in both governmental and non-governmental facilities. For example, a reliable supply of running water and electricity was available in only 68 and 64 % of facilities, respectively. Oxygen was unavailable at one in four facilities. Uninterrupted access to plain radiographs and materials for splinting and casting were available in only 27 and 48 % of facilities with ≤100 beds, respectively, versus 90 and 66 % of facilities with >300 beds. Sterile gloves were available in three of four facilities overall, and in only four of five facilities with >300 beds. The information on human resources is expressed as the average number of providers per facility, and indicated the paucity of trained surgeons at the primary referral level. Task shifting was utilized at all levels of health facility for surgical care, but was the predominant strategy utilized at smaller, primary level facilities.

Table 2 Availability of infrastructure, supplies, and human resources
Full size table

The availability of selected surgical and anesthetic services for the entire group, type of facility, and number of beds is shown in Table 3. Wound debridement and drainage of abscesses are reliably performed at the majority of facilities. In contrast, less than 50 % of primary referral facilities or those facilities with less than 100 beds were able to perform any of the orthopedic interventions. Similarly, Cesarean section and laparotomy were available in only 52 and 42 % of those will less than 100 beds, respectively, and similar findings were observed with the availability of anesthetic services.

Table 3 Availability of orthopedic and anesthesia services
Full size table

Our statistical analysis revealed that for facilities with ≤100 beds, availability of infrastructure, physical resources and supplies, and surgical and anesthetic interventions was significantly lower at public versus Private/NGO/Mission facilities for every item except a blood bank and abscess drainage (Table 4). Similar findings were observed in facilities with 100–300 beds for the availability of the following: electricity, postoperative care unit, blood bank, guidelines for surgical care, sterilizer, sterile gloves, splints/casts, and regional anesthesia. No significant differences were found when comparing facilities with more than 300 beds.

Table 4 Uninterrupted availability of infrastructure, physical resources and supplies, human resources, and interventions
Full size table

Discussion

The burden of acute musculoskeletal conditions in LMICs is substantial, and the available evidence suggests that considerable morbidity can be averted by the provision of safe and timely orthopedic services, especially for injuries and infections. Ninety percent of the more than 5 million mortal injuries each year occur in LMICs [1, 5], and for each death, there may be one case of permanent disability [3, 4]. Findings from the most recent iteration of the global burden of disease study indicate that injury-related years lived with disability (YLD’s) have increased by 6.5 %, and it is projected that YLD’s will increase by 38.4 % by 2030 [6]. Improvements in trauma care will likely increase the volume of non-fatal injuries, elevating the importance of orthopedic care. Acute osteomyelitis is most commonly due to hematogenous seeding, or as a complication of open fractures or surgical procedures on bone. The incidence of acute hematogenous osteomyelitis (AHO) varies from 43 to 200 cases per 100,000 persons in developing countries [39]. We are unaware of any reliable figures concerning the incidence of open fractures or surgical site infections in LMICs, and the burden of musculoskeletal infections (osteomyelitis and septic arthritis) has not been quantified with existing metrics.

While there is evidence to suggest that approximately 15 % of the world’s population is living with a disability [40], information concerning that component due to musculoskeletal causes is limited and is mainly experiential. There are 20–40 million non-fatal injuries each year from road traffic crashes alone [2, 5], and the reported prevalence of disability varies from 2 to 87 % [7]. Mock et al. found that 0.83 % of Ghanaians were disabled because of an injury, 78 % of which involved the extremities [8]. Atijosen et al. estimated that 5.2 % of the Rwandan population was disabled due to a musculoskeletal condition, including injuries (31 %) and infections (4 %) [9]. Based on these two studies, assuming a 1.1 % prevalence of permanent disability, we would estimate that more than 64 million people in LMICs may be permanently disabled because of an acute musculoskeletal condition.

Our study revealed deficiencies in the availability of infrastructure, equipment and supplies, and human resources required to deliver essential orthopedic surgical services, especially at smaller or primary referral level facilities. These observations support those of previous studies using the GIEESC tool (10), the tool developed by the Harvard Humanitarian Initiative (5), the PIPES tool (2), and demographic health surveys/service provision assessments (Hsia). These reports have also involved a convenience sample of facilities, public and private, from the primary referral level through the tertiary level, and selected data are shown in Table 5. We also found that for most data points, availability was lower in public facilities versus Private/NGO/Mission facilities for those facilities with ≤100 beds, and for a smaller number of indicators in facilities with 100–300 beds.

Table 5 Review of selected items of infrastructure for surgical care and musculoskeletal surgical interventions
Full size table

The information on infrastructure is germane to the delivery of any facility-based health services, medical or surgical. Previous studies have noted that an uninterrupted supply of water was available in 18–100 % [12–20, 22–27, 29], electricity in 48–89 % [13–20, 22–27, 29], and oxygen in 28–100 % of facilities [11–20, 22–27, 29] (Table 5). Frequent power outages were noted in Bangladesh, Uganda, and Rwanda. The costs of electricity and/or diesel fuel for a generator may also be a challenge. While a blood bank was always available on-site in 23–64 % [12–15, 19, 23–27, 29], versus 13–84 % in our study, some facilities have access to local or regional blood banks, or can process on-site, immediate donations. Only five studies commented on the availability of plain radiographs (0–44 %) [15, 18, 19, 23, 29]. Hsia et al. studied surgical care in five African countries (Kenya, Tanzania, Rwanda, Uganda, and Ghana) and noted deficiencies in infrastructure, equipment, medical storage, infection control, education, and quality control [30]. The authors found that less than 50 % of facilities had the capacity to repair or maintain equipment [30].

With regard to equipment and supplies, a sterilizer was present in 41–100 % of facilities, and sterile gloves in only 52–90 % of facilities [11, 13, 15–19, 21, 26, 29], similar to the findings of our study. Materials for splinting and casting were available in 14–85 % in the five studies in which this was measured [13, 19, 21, 23, 29]. Only two studies reported whether or not a tourniquet was available, with values ranging from 30 to 79 %, versus 61–86 % in the present study.

The availability of these musculoskeletal interventions varied considerably in previous studies utilizing the GIEESC tool. Simple wound care and irrigation and debridement of abscesses were available at the majority of facilities. Closed management of fractures was performed in 30–100 % [11, 13–16, 19–23, 29], open management of fractures in 6–100 % [11, 13, 14, 17, 19–23, 29], treatment of joint dislocations in 43–100 % [13, 14, 19–21, 29], amputations in 39–100 % [11, 13–17, 19–23, 29], and drainage of osteomyelitis or septic arthritis in 32–100 % [11, 13, 14, 17, 19–23, 29] (Table 5). Significant deficiencies were noted at the primary referral level or in facilities with <100 beds. In Malawi, orthopedic procedures are typically performed at the central hospitals, and procedures for fracture care and osteomyelitis accounted for only 6 and 2 % of the volume at district hospitals, respectively [31]. While delayed management of acute musculoskeletal conditions often results in disability, a lack of timely access to other essential surgical services such as Cesarean section, which was available at 41–96 % of facilities in our literature review, leads to countless unnecessary deaths of mothers and infants. There are also gross deficiencies in the number of surgeons in LMICs, especially at primary referral level facilities, and the majority practice at tertiary facilities in urban centers, often in the private sector. The majority of surgical services at the primary referral level are provided by general surgeons, or by medical doctors and/or paraprofessionals [11–16, 18–20, 22–29], such as the Orthopedic Clinical Officers (Malawi) [41] or Clinical Officers (Uganda) [26]. The limited information available concerning orthopedic subspecialists suggests that there are approximately 9 orthopedic surgeons in Rwanda [9, 23], 9 in Malawi [41], and 24 in Ghana [42], to care for more than 51 million people [43]. In contrast, Lebrun et al. found that there were an average of 1.1 orthopedic surgeons at each district facility and 5.3 at each medical college in Bangladesh [25].

We must also recognize that anesthetic services are an essential component of surgical care. An anesthesia machine was present in only 32–100 % of facilities in our literature review [14, 15, 17–21, 23–25], and the percentages of facilities offering selected anesthesia services were as follows: general (25–72 %), spinal (42–100 %), ketamine (44–100 %), and regional (18–100 %). A previous review from the WHO database involving 590 facilities in 22 countries found that general anesthesia was available in 59 %, spinal in 66 %, Ketamine in 72 %, and regional in 56 % [44].

There are a number of limitations that must be mentioned. The data have accumulated gradually over nearly 7 years, and no formal sampling methodology has been utilized. We might view the results as a “best case scenario,” considering our findings were from a convenience sample. As noted previously, we recognize that there is a degree of overlap between number of beds, and how each facility chose to classify itself based on the choices available on the questionnaire, resulting in some lack of consistency. For example, while the majority of primary health facilities are smaller facilities with less than 100 beds, a small number of such facilities had more than 300 beds (Fig. 1). In addition, the findings may not accurately reflect an ever-changing landscape in which there are interruptions to the supply chain, where maintenance of equipment is variable and replacement is often delayed or not possible, and where the number and skill of health workers may be in a constant state of flux. The surgical workforce may intermittently be supplemented by surgical providers from other levels of the health system, or by health workers from NGOs or other organizations. We recognize that expectations for service delivery at each tier in a system may vary between and within countries, and is some degree of overlap based on the levels of analysis that we have selected. The surgical situational analysis tool has been shown to have high reliability on structure, but poor reliability on process of care [45]. The WHO tool was intended to inform improvements in service delivery at the individual facilities level. While we can state what percentage of facilities offered a particular service, we cannot draw any conclusions concerning the quantity or quality of services delivered, or on patient outcomes. In addition, the GIEESC tool was designed for primary level health facilities and lists only “surgeon” without the opportunity to indicate whether that provider is a subspecialist. As such, we cannot draw any conclusions about the number of orthopedic surgeons or where they practice. Finally, our data are just a snapshot, and are insufficient to inform policy changes. In addition, there is great need to integrate a monitoring strategy for service availability within each countries health information system (HIS) to improve service availability, and ultimately improve service delivery. The information collected must be tailored to specific levels within the HIS, for example, managers at individual health facilities versus health planners at the regional or national level, so questionnaires must be adapted. Monitoring tools must also be developed, for example, elements of the WHO situational analysis have been integrated into WHO’s service availability and readiness assessment (SARA) [46]. While future iterations of the GIEESC tool will likely be of greatest benefit to managers at the facilities and perhaps regional level, the tool must be adapted to accurately capture process measures to ensure that safety and quality services are monitored.

Conclusion

Deficiencies in the availability of orthopedic surgical services, as well as life-saving procedures such as Cesarean section and laparotomy, were observed at all levels of health facility in this group of low- and lower-middle-income countries. These observations were most pronounced in facilities with <100 beds, especially in the public sector. Given that a majority of patients in low- and lower-middle-income countries receive their health care services at such smaller, public facilities, strengthening the delivery of surgical services including orthopedics must be prioritized if we are to reduce the burden of death and disability from a variety of emergent health conditions.