The Laryngoscope C 2014 The American Laryngological, V
Rhinological and Otological Society, Inc.
Cleft Lip–Cleft Palate in Zimbabwe: Estimating the Distribution of the Surgical Burden of Disease Using Geographic Information Systems Travis T. Tollefson, MD, MPH; David Shaye, MD; Blythe Durbin-Johnson, PhD; Omid Mehdezadeh, MD; Leonard Mahomva, BDS; Midion Chidzonga, BDS, FFD Objectives/Hypothesis: To evaluate the prevalence and unmet need for cleft lip–cleft palate reconstructive surgery by using incidence. Our hypotheses were that the age of presentation to screening clinics will decrease between 2006 and 2012, and the geospatial distribution of cases will expand to a more rural catchment area. Study Design: Longitudinal cross-sectional/geospatial distribution study. Methods: An online, secure database was created from intake forms for children with cleft lip–cleft palate (N 5 604) in Zimbabwe (2006–2012). Univariate analysis was completed. A linear regression model was fitted to test the time trend of a child’s age at the time of presentation. Unique patient addresses (n 5 411) were matched. Maps presenting cleft diagnosis and presentation year were created with geographic information systems (GIS) software. Results: The median age of presentation was greater for isolated cleft palate (4.2 years, n 5 106) than isolated cleft lip (1.5 years, n 5 251) and cleft lip–cleft palate (2.0 years, n 5 175). Cleft lip cases were mostly left sided with equal gender distribution. The overall age of presentation remained stable (P 5.83). The age of children with isolated cleft palate decreased by 0.8 years per surgical trip (P 5.01), suggesting the prevalence of unrepaired cleft palate is decreasing due to local and visiting surgeons. The catchment area extended to a less populous area, but clustered around Harare and Bulawayo. Conclusions: This study gives Zimbabwe-specific evidence that supports reports of the persistent burden of disease requiring attention. The GIS software provided data for the primary needs assessment, which will direct communication to healthcare providers and prospective patients outside of the current catchment area. Key Words: Cleft lip, cleft palate, global surgery, burden of disease, low- and middle-income countries, developing country, geospatial analysis, unmet surgical need.. Level of Evidence: 3. Laryngoscope, 125:S1–S14, 2015
Additional Supporting Information may be found in the online version of this article. From the Department of Otolaryngology–Head and Neck Surgery, (T.T.T), University of California, Davis, Sacramento, California, U.S.A; Department of Biostatistics (B.D.-J.), University of California, Davis, Sacramento, California, U.S.A; Department of Otolaryngology (D.S.), University of Minnesota, Minneapolis, Minnesota, U.S.A; Department of Otolaryngology-Head and Neck Surgery, New York University–Langone Medical School (O.M.), New York, New York; and Department of Oral and Maxillofacial Surgery (L.M., M.C.), University of Zimbabwe College of Health Sciences, Harare, Zimbabwe Editor’s Note: This Manuscript was accepted for publication April 29, 2014. The authors have no funding, financial relationships, or conflicts of interest to disclose. The statistical analysis was supported by the National Center for Advancing Tanslational Sciences, National Institutions of Health, through grant UL1 TR000002. The authors would like to acknowledge “Jennifer Trubenbach, Joseph Clawson, MD, and Operation of Hope Team Members; Este Geraghty, MD (UC Davis) and Jeff Blossum (Harvard School of Public Health) for GIS support; and E. Francis Cook, Jr, SCD and Heather Baer, SCD at the Harvard School of Public Health for serving as mentors in the project’s initiation.” Send correspondence to Travis T. Tollefson, MD, Associate Professor, Facial Plastic and Reconstructive Surgery, Department of Otolaryngology–Head and Neck Surgery, University of California, Davis Medical Center, 2521 Stockton Blvd., Suite 7200, Sacramento, CA 95817. E-mail: [email protected] DOI: 10.1002/lary.24747
Laryngoscope 125: February 2015
The human desire to lighten a little the torments of all these poor. . .creates a kind of energy which gives one a positive craving to relieve as many as one can. Henri Dunant, Founder, International Committee of the Red Cross, Geneva, Switzerland, 1863.
INTRODUCTION Shift to Global Surgery Cleft lip and cleft palate can be devastating facial deformities that affect a child’s nutrition, speech, hearing, and quality of life. Early surgical treatment and interdisciplinary care can avert some of this disability. In 2006, we began a recurring cleft lip–cleft palate program in Zimbabwe with a team of five to more than two dozen United States-based healthcare providers and numerous hostcountry physicians and volunteers. The methodology has evolved from concentrating on providing only surgical care (vertical model) toward a more diagonal model of care delivery. Vertical, horizontal, and diagonal models of care delivery have been defined. The short-term surgical teams, which provide services with minimal interaction with the local healthcare system, are an example of the vertical model. The horizontal model emphasizes working within the infrastructure of the host-country’s healthcare system. Tollefson et al.: Cleft Lip–Cleft Palate in Zimbabwe
S1
The diagonal model emphasizes integration of external and international surgical services with existing healthcare infrastructure through education, improving quality of care, and increasing surgical capacity (see Appendix A1 for further discussion). In this article, we present the first phase, which is a needs assessment of the burden of surgical disease attributed to cleft lip–cleft palate in Zimbabwe by estimating the prevalence and geographic distribution of patients with cleft lip–cleft palate in Zimbabwe. Global public health efforts and funding are prioritized to address and prevent premature death and disability, known as the burden of disease (Appendix A2). Traditionally, infectious diseases like diarrheal illness, malaria, tuberculosis, or human immunodeficiency virus (HIV) have been the focus of resources and funding. Policy makers and funding agencies are increasingly adding surgical services to global health initiatives for these resource-poor settings, but require evidence of the unmet need potential impact.1,2 We will present a needs assessment (prevalence and geographic distribution) of the surgical burden of cleft lip–cleft palate in Zimbabwe, along with an increasing body of literature in global surgery (Appendix A3). The developing field of global surgery has emerged from a variety of stakeholders who support surgical services in low- and middle-income countries (LMICs).3,4 International aid agencies, nongovernment organizations (NGOs), national healthcare systems, and academic institutions have partnered to answer the difficult questions of how to provide surgical services in countries where funding, resources, manpower, and often education are limited. In LMICs, a significant proportion of cleft lip–cleft palate care is provided by visiting surgical teams, which contributes to the estimated $250 million/year spent by NGOs.5 The model for providing surgical services in resource-poor settings is shifting from emphasis on the volume of cases performed to sustainability, education with host-country providers, and health outcomes measurement of the entire care process.1,6 The relative paucity of quality research data limits our understanding of the actual magnitude of the yet to be addressed surgical needs in these resource-poor settings.3,7,8 Research priorities regarding the global burden of surgical disease are focused on understanding how much surgical disease is untreated and where is it most prevalent (Appendix A2). These priorities can be divided into the following: 1) to establish the met/unmet need for surgical services by data collection, which this study addresses; 2) to find to what extent surgical repair prevents lifelong disability; 3) to determine benchmarks for the quality of surgical care and its delivery; and 4) to explore the cost-effectiveness of surgical delivery.1,6,9 The long-term goal of our project team was to use an evidence-based approach to foster an integrative, or diagonal, model of surgical care delivery (Appendix A1) within the Zimbabwean healthcare system. This 7-year longitudinal, cross-sectional assessment will be presented with a brief introduction to the following: 1) the burden of surgical disease, its components, research objectives, and metrics for measurement; and 2) the rationale for choosing Zimbabwe to implement the program.
Burden of Disease and Surgery Surgery, as a necessary component to address the global burden of disease, has only recently been Laryngoscope 125: February 2015
S2
acknowledged by global health policy makers and funding agencies for projects in resource-poor settings. In an effort to measure worldwide diseases and injuries, the World Bank commissioned a preliminary burden of disease study in 1991.10 The Disease Control Priorities Project (2nd edition) crudely estimated that 11% of premature death or disability could be averted with surgical services for trauma, oncologic, and congenital deformities (e.g., cardiac defects, cleft lip and palate, and club foot).10,11 The World Health Organization (WHO) partnered with academic institutions to update this 2010 Global Burden of Diseases, Injuries, and Risk Factors Study with country-specific disease burden, life expectancy, and years of healthy life lost due to disability (Appendix A2).12 To clarify some definitions, surgical conditions are “any disease state requiring the expertise of a surgically trained provider,”8 that requires anesthesia for incision, excision, and suture.9 The burden of surgical conditions is therefore “the total disability and premature deaths that would occur in a population should there be no surgical care.”8,10
Healthcare Infrastructure in Zimbabwe The choice of Zimbabwe for this cleft lip–cleft palate project may seem counterintuitive, but simply stated, there is a great need and little international support for surgical services. Infectious disease (HIV/acquired immune deficiency syndrome [AIDS], tuberculosis, malaria, and cholera) and maternal–child mortality prevention are the priorities for international public health support in Zimbabwe.13 One could ask if the surgical repair of congenital facial deformities is even meaningful in the setting of severe economic crisis, cholera outbreaks, and massive burden of infectious diseases (Appendix A4). We believe that a diagonal model (Appendix A1) integrated into the ministry of health through a short-term surgical team and iterative quality and capacity improvement is worthwhile. The healthcare system in Zimbabwe has a rich tradition of strong allied health schools and experience with integrating with international NGOs for HIV management. This would suggest that educational efforts could be effective and promote future self-sufficiency. There are many overlapping, international recurring short-term cleft surgical teams that volunteer throughout South and Central America, Asian, and Africa, but few are active in Zimbabwe. The southern African country of Zimbabwe is a landlocked country with a rich geopolitical and socioeconomic history. Harare, located in the northeast of the country, is the capital of the 13.1-million population of the country. Harare has an estimated population of 2 million, but this is a fluid number as high unemployment rates have driven an influx of people looking for work from rural areas. Events in the 1990s led to catastrophic economic and political changes that resulted in the highest inflation rate (>11 million percent) in the world in 200814 and a persistently high unemployment rate. This unrest was compounded by a high prevalence of HIV Tollefson et al.: Cleft Lip–Cleft Palate in Zimbabwe
infection and other infectious diseases, contributing to a life expectancy of below 40 years of age. During this period, the workforce lost a significant number of educated professionals and physicians due to emigration. The number of physicians in 2003 was noted to be approximately 1:10,000 (with even fewer subspecialist surgeons), in comparison to the United States with 19 physicians per 10,00013 (see Appendix A3 for additional geography, history, and healthcare indices of Zimbabwe). The prevalence of cleft lip–cleft palate in Zimbabwe has not been well documented. The incidence of cleft lip–cleft palate in the United States is often quoted to be approximately 1:600 to 1:750 live births15 (Appendix A4). Intermediate rates are seen in Caucasians (1:1,000 live births), whereas the lowest rates are seen within the African American community (0.3:1,000 live births).16–19 To relate this to African incidence rates, the incidence in Malawi has been described as 0.67:1,000 live births of cleft lip–cleft palate, but reports from Zimbabwe have been sparse.19 Therefore, to extrapolate from existing data, if the population of Zimbabwe is between 12 and 13.1 million in 2011, with a live birth rate of 27.7 live births per 1,000,20 and if we use the incident rate from the Malawi data of 0.67:1000, we would find approximately 217 live births with cleft lip–cleft palate per year. Even if the US rate of 1.67/1,000 live births were used, only 272 children with cleft lip–cleft palate would be born a year in Zimbabwe, a number that can reasonably be addressed. Of those live births, children with oral clefts, when compared to the overall population, have twice the odds of dying within the first year of life, which may be averted by implementing early interventions such as feeding and surgical services.21
Study Objectives The principle objective of this study was to evaluate the prevalence and unmet need for cleft lip–cleft palate reconstructive surgery by measuring the age of patient presentation. The incidence of cleft lip–cleft palate, acting as a proxy, can be used to estimate the unmet surgical burden of other congenital deformities. Our general hypothesis is that there remains a significant unmet need for surgical correction of congenital deformities. Evaluating the age of patient presentation may act as a proxy for overall disease burden, in that older patients, once treated, cause the age of presentation to decrease as the backlog of the surgical burden is managed. The older patients could be considered to have been treated by our team or related local surgeons, but other visiting teams were not present or very rare during the study period. A decrease in the overall burden may demonstrate a temporal trend in patient age at presentation. Our hypothesis was that the age of presentation to the biannual cleft team consultations will decrease in the overall patient population between 2006 and 2012. The secondary hypothesis was that the geospatial distribution of cleft lip–cleft palate may demonstrate clustering or expansion of catchment area to more rural areas (possibly due to expansion of communication about Laryngoscope 125: February 2015
the program, news coverage, and partnering). The research question asked is that compared to those children with isolated cleft palate, are those with cleft lip more likely to be presented to screening clinics due to the visibility of the cleft lip, as opposed to the less obvious palate defect? In the more rural areas, healthcare providers may not evaluate the palate as readily as an obvious cleft lip. Using geographic information systems (GIS) software, we sought to create epidemiological surveillance of where the children presenting to the cleft team screening clinics were travelling.
METHODS Study Design Two study designs are nested in this longitudinal crosssectional study. First, the epidemiologic aspect relies on estimating the prevalence of cleft lip–cleft palate from data on incidence and duration (age of patient). The second component involves estimating the geographic distribution of cleft lip–cleft palate prevalence in Zimbabwe for children presenting for evaluation to biannual surgery screening clinics in Harare, Zimbabwe between 2006 and 1012. Since 2006, the surgical team, Operation of Hope, has been making biannual surgical missions to evaluate, triage, and surgically correct cleft lip–cleft palate in over 600 patients. On the day of clinic presentation, the study sample was enrolled into the screening clinic using a standardized screening form (Fig. 1).
Database Creation: Study Population The screening form data were rigorously collected and entered into a secure, online database (Velos eResearch; Velos, Inc., Fremont, CA), which had been designed in conjunction with the University of California, Davis Biomedical Informatics, Clinical & Translational Sciences Center. Institutional review board approval was obtained through both the University of Zimbabwe School of Health Sciences and the University of California, Davis for this study. This method of capture is limited by including only those children who present for clinic evaluation.
Variables Study variables were based on the screening clinic forms. Our outcome measures included: continuous explanatory variables included age, birth weight, current weight, hemoglobin level, and travel distance in hours. Categorical variables were gender, diagnosis (isolated cleft lip, isolated cleft palate, cleft lip–cleft palate), diagnosis laterality, type of procedure, and revision surgery. Dichotomous variables included self-reported HIV status and tuberculosis status. The spatial variables were the home address, city, and province in Zimbabwe, which were converted to longitude and latitude.
Statistical Analysis Univariate analyses were conducted for all continuous variables to examine measures of central tendency using the Students t test and v2 test for categorical variables. Linear regression modeling was used to test for a time trend in patient age. The dependent variable was age of patient presentation, and the independent variable was the sequential year of recurring surgical trip. Analyses were conducted in R, version 2.13.0 (R Development Core Team, 2011; The R Foundation for
Tollefson et al.: Cleft Lip–Cleft Palate in Zimbabwe
S3
Fig. 1. Data entry form for each screening clinic for creation of the database (n 5 604).
Statistical Computing, Vienna, Austria). Statistical significance was considered as P .05.
Spatial Analysis: Application of Database Data to GIS Mapping Tool The Zimbabwe cleft lip–cleft palate data from 2006 to 2010 were incorporated into a spreadsheet format (Excel 2003; Microsoft, Redmond, WA). These data were converted for us in the GIS software (ArcInfo 10.0; ESRI, Redlands, CA). Geocoding was completed to obtain latitude and longitude coordinates for each patient’s home address based on the individual cities, towns, and closest village using Google Earth 7.1.1.1888. (Google Earth version 5.1.3533.1731, 2009; Google Inc., Mountain View, CA). Initially, 450 records were matched, but errors were discovered during the rematching. Abbreviated city names or provinces, misspellings, and street names were reviewed and corrected where possible. A total of 411 patients with cleft lip– cleft palate were captured. These data were used to construct maps for each year of presentation from 2006 to 2010 for all patients with a cleft (n 5 411). The geographic distribution of cases of cleft lip–cleft palate was created in sequential surgical team “mission/camp/brigade” based on month and year of the team arrival. For comparison, available population density maps were used as reference for where the largest population base is located in the country. Separate diagnosis-specific maps were created from longitude and latitude coordinates (cleft lip–cleft palate, isolated cleft lip, and isolated cleft palate) (Fig. 1). The distances and estimated travel times from home to the Harare screening clinics were calculated for the six most common locations.
Laryngoscope 125: February 2015
S4
RESULTS Patient characteristics are listed in Table I. Of the total 604 patients, 175 represented the combined cleft lip–cleft palate group. Isolated cleft palate was seen in 106 patients and isolated cleft lip in 251 patients. The median age for the cleft lip and cleft lip–cleft palate was between 1.5 and 2.0 years, whereas the isolated cleft palate patient’s median age at presentation was 4.2 years. The patient’s gender was evenly distributed in the cleft lip with and without cleft palate groups, and no difference was identified between the diagnosis groups (P 5.81). A slight male preponderance was noted in the isolated cleft palate group, with 50% male, 42% female, and 8% of unknown gender. Birth weight was significantly less for those with isolated cleft palate at 2.8 kg (standard deviation [SD] 0.5) compared to isolated cleft lip at 3.0 kg (SD 0.9) and cleft lip–cleft palate at 2.9 kg (SD 0.5). This suggestion of a lower birth weight for those with isolated cleft palate is still within the normal range and likely does not represent a significant clinical difference. The screening process only identified three patients with tuberculosis, and five were being treated for HIV. Preoperative hemoglobin levels (total mean 11.5 g/dL) were not different between diagnosis (P 5.398). The left-sided preponderance of cleft lip was seen to be present, with approximately 42% to 47% of the patients demonstrating a left-sided cleft lip and 23% to 34% as a right-sided cleft lip (Figs. 2 and 3). Tollefson et al.: Cleft Lip–Cleft Palate in Zimbabwe
TABLE I. Patient Characteristics by Diagnosis. Variable
Cleft Lip, n 5 251
Cleft Palate, n 5 106
Cleft Lip and Palate, n 5 175
4.9 (6.7) 1.5 (0.0–43)
5.5 (5.0) 4.2 (0.6–33)
4.9 (6.3) 2.0 (0.0–34)
3
—
1
Age, yr
No. missing Gender Male
6.3 (8.6) 3.0 (0.04–80) 5 .081
Female Unknown Birth weight, kg
122 (49%)
47 (44%)
82 (47%)
288 (48%)
118 (47%)
52 (49%)
90 (51%)
291 (48%)
11 (4%)
7 (7%)
3 (2%)
25 (4%) .025
Mean (SD)
3.1 (0.9)
2.8 (0.5)
2.9 (0.5)
3.0 (0.7)
3.0 (1.3–9.2) 117
2.9 (1.8–3.8) 47
3.0 (1.2–4.0) 39
3.0 (1.2–4.1) 247
16.6 (15.2) 10.0 (3.0–78)
17.2 (10.9) 13.8(4.6–62)
15.8 (13.9) 10.5 (2.9–70)
46
24
6
Current weight, kg Mean (SD) Median (range)
All Patients, N 5 604†
.250
Mean (SD) Median (range)
Median (range) No. missing
P Value*
.501
No. missing Diagnosis description Left
18.2 (15.9) 12.0(2.9–78) 95
Rhinological and Otological Society, Inc.
Cleft Lip–Cleft Palate in Zimbabwe: Estimating the Distribution of the Surgical Burden of Disease Using Geographic Information Systems Travis T. Tollefson, MD, MPH; David Shaye, MD; Blythe Durbin-Johnson, PhD; Omid Mehdezadeh, MD; Leonard Mahomva, BDS; Midion Chidzonga, BDS, FFD Objectives/Hypothesis: To evaluate the prevalence and unmet need for cleft lip–cleft palate reconstructive surgery by using incidence. Our hypotheses were that the age of presentation to screening clinics will decrease between 2006 and 2012, and the geospatial distribution of cases will expand to a more rural catchment area. Study Design: Longitudinal cross-sectional/geospatial distribution study. Methods: An online, secure database was created from intake forms for children with cleft lip–cleft palate (N 5 604) in Zimbabwe (2006–2012). Univariate analysis was completed. A linear regression model was fitted to test the time trend of a child’s age at the time of presentation. Unique patient addresses (n 5 411) were matched. Maps presenting cleft diagnosis and presentation year were created with geographic information systems (GIS) software. Results: The median age of presentation was greater for isolated cleft palate (4.2 years, n 5 106) than isolated cleft lip (1.5 years, n 5 251) and cleft lip–cleft palate (2.0 years, n 5 175). Cleft lip cases were mostly left sided with equal gender distribution. The overall age of presentation remained stable (P 5.83). The age of children with isolated cleft palate decreased by 0.8 years per surgical trip (P 5.01), suggesting the prevalence of unrepaired cleft palate is decreasing due to local and visiting surgeons. The catchment area extended to a less populous area, but clustered around Harare and Bulawayo. Conclusions: This study gives Zimbabwe-specific evidence that supports reports of the persistent burden of disease requiring attention. The GIS software provided data for the primary needs assessment, which will direct communication to healthcare providers and prospective patients outside of the current catchment area. Key Words: Cleft lip, cleft palate, global surgery, burden of disease, low- and middle-income countries, developing country, geospatial analysis, unmet surgical need.. Level of Evidence: 3. Laryngoscope, 125:S1–S14, 2015
Additional Supporting Information may be found in the online version of this article. From the Department of Otolaryngology–Head and Neck Surgery, (T.T.T), University of California, Davis, Sacramento, California, U.S.A; Department of Biostatistics (B.D.-J.), University of California, Davis, Sacramento, California, U.S.A; Department of Otolaryngology (D.S.), University of Minnesota, Minneapolis, Minnesota, U.S.A; Department of Otolaryngology-Head and Neck Surgery, New York University–Langone Medical School (O.M.), New York, New York; and Department of Oral and Maxillofacial Surgery (L.M., M.C.), University of Zimbabwe College of Health Sciences, Harare, Zimbabwe Editor’s Note: This Manuscript was accepted for publication April 29, 2014. The authors have no funding, financial relationships, or conflicts of interest to disclose. The statistical analysis was supported by the National Center for Advancing Tanslational Sciences, National Institutions of Health, through grant UL1 TR000002. The authors would like to acknowledge “Jennifer Trubenbach, Joseph Clawson, MD, and Operation of Hope Team Members; Este Geraghty, MD (UC Davis) and Jeff Blossum (Harvard School of Public Health) for GIS support; and E. Francis Cook, Jr, SCD and Heather Baer, SCD at the Harvard School of Public Health for serving as mentors in the project’s initiation.” Send correspondence to Travis T. Tollefson, MD, Associate Professor, Facial Plastic and Reconstructive Surgery, Department of Otolaryngology–Head and Neck Surgery, University of California, Davis Medical Center, 2521 Stockton Blvd., Suite 7200, Sacramento, CA 95817. E-mail: [email protected] DOI: 10.1002/lary.24747
Laryngoscope 125: February 2015
The human desire to lighten a little the torments of all these poor. . .creates a kind of energy which gives one a positive craving to relieve as many as one can. Henri Dunant, Founder, International Committee of the Red Cross, Geneva, Switzerland, 1863.
INTRODUCTION Shift to Global Surgery Cleft lip and cleft palate can be devastating facial deformities that affect a child’s nutrition, speech, hearing, and quality of life. Early surgical treatment and interdisciplinary care can avert some of this disability. In 2006, we began a recurring cleft lip–cleft palate program in Zimbabwe with a team of five to more than two dozen United States-based healthcare providers and numerous hostcountry physicians and volunteers. The methodology has evolved from concentrating on providing only surgical care (vertical model) toward a more diagonal model of care delivery. Vertical, horizontal, and diagonal models of care delivery have been defined. The short-term surgical teams, which provide services with minimal interaction with the local healthcare system, are an example of the vertical model. The horizontal model emphasizes working within the infrastructure of the host-country’s healthcare system. Tollefson et al.: Cleft Lip–Cleft Palate in Zimbabwe
S1
The diagonal model emphasizes integration of external and international surgical services with existing healthcare infrastructure through education, improving quality of care, and increasing surgical capacity (see Appendix A1 for further discussion). In this article, we present the first phase, which is a needs assessment of the burden of surgical disease attributed to cleft lip–cleft palate in Zimbabwe by estimating the prevalence and geographic distribution of patients with cleft lip–cleft palate in Zimbabwe. Global public health efforts and funding are prioritized to address and prevent premature death and disability, known as the burden of disease (Appendix A2). Traditionally, infectious diseases like diarrheal illness, malaria, tuberculosis, or human immunodeficiency virus (HIV) have been the focus of resources and funding. Policy makers and funding agencies are increasingly adding surgical services to global health initiatives for these resource-poor settings, but require evidence of the unmet need potential impact.1,2 We will present a needs assessment (prevalence and geographic distribution) of the surgical burden of cleft lip–cleft palate in Zimbabwe, along with an increasing body of literature in global surgery (Appendix A3). The developing field of global surgery has emerged from a variety of stakeholders who support surgical services in low- and middle-income countries (LMICs).3,4 International aid agencies, nongovernment organizations (NGOs), national healthcare systems, and academic institutions have partnered to answer the difficult questions of how to provide surgical services in countries where funding, resources, manpower, and often education are limited. In LMICs, a significant proportion of cleft lip–cleft palate care is provided by visiting surgical teams, which contributes to the estimated $250 million/year spent by NGOs.5 The model for providing surgical services in resource-poor settings is shifting from emphasis on the volume of cases performed to sustainability, education with host-country providers, and health outcomes measurement of the entire care process.1,6 The relative paucity of quality research data limits our understanding of the actual magnitude of the yet to be addressed surgical needs in these resource-poor settings.3,7,8 Research priorities regarding the global burden of surgical disease are focused on understanding how much surgical disease is untreated and where is it most prevalent (Appendix A2). These priorities can be divided into the following: 1) to establish the met/unmet need for surgical services by data collection, which this study addresses; 2) to find to what extent surgical repair prevents lifelong disability; 3) to determine benchmarks for the quality of surgical care and its delivery; and 4) to explore the cost-effectiveness of surgical delivery.1,6,9 The long-term goal of our project team was to use an evidence-based approach to foster an integrative, or diagonal, model of surgical care delivery (Appendix A1) within the Zimbabwean healthcare system. This 7-year longitudinal, cross-sectional assessment will be presented with a brief introduction to the following: 1) the burden of surgical disease, its components, research objectives, and metrics for measurement; and 2) the rationale for choosing Zimbabwe to implement the program.
Burden of Disease and Surgery Surgery, as a necessary component to address the global burden of disease, has only recently been Laryngoscope 125: February 2015
S2
acknowledged by global health policy makers and funding agencies for projects in resource-poor settings. In an effort to measure worldwide diseases and injuries, the World Bank commissioned a preliminary burden of disease study in 1991.10 The Disease Control Priorities Project (2nd edition) crudely estimated that 11% of premature death or disability could be averted with surgical services for trauma, oncologic, and congenital deformities (e.g., cardiac defects, cleft lip and palate, and club foot).10,11 The World Health Organization (WHO) partnered with academic institutions to update this 2010 Global Burden of Diseases, Injuries, and Risk Factors Study with country-specific disease burden, life expectancy, and years of healthy life lost due to disability (Appendix A2).12 To clarify some definitions, surgical conditions are “any disease state requiring the expertise of a surgically trained provider,”8 that requires anesthesia for incision, excision, and suture.9 The burden of surgical conditions is therefore “the total disability and premature deaths that would occur in a population should there be no surgical care.”8,10
Healthcare Infrastructure in Zimbabwe The choice of Zimbabwe for this cleft lip–cleft palate project may seem counterintuitive, but simply stated, there is a great need and little international support for surgical services. Infectious disease (HIV/acquired immune deficiency syndrome [AIDS], tuberculosis, malaria, and cholera) and maternal–child mortality prevention are the priorities for international public health support in Zimbabwe.13 One could ask if the surgical repair of congenital facial deformities is even meaningful in the setting of severe economic crisis, cholera outbreaks, and massive burden of infectious diseases (Appendix A4). We believe that a diagonal model (Appendix A1) integrated into the ministry of health through a short-term surgical team and iterative quality and capacity improvement is worthwhile. The healthcare system in Zimbabwe has a rich tradition of strong allied health schools and experience with integrating with international NGOs for HIV management. This would suggest that educational efforts could be effective and promote future self-sufficiency. There are many overlapping, international recurring short-term cleft surgical teams that volunteer throughout South and Central America, Asian, and Africa, but few are active in Zimbabwe. The southern African country of Zimbabwe is a landlocked country with a rich geopolitical and socioeconomic history. Harare, located in the northeast of the country, is the capital of the 13.1-million population of the country. Harare has an estimated population of 2 million, but this is a fluid number as high unemployment rates have driven an influx of people looking for work from rural areas. Events in the 1990s led to catastrophic economic and political changes that resulted in the highest inflation rate (>11 million percent) in the world in 200814 and a persistently high unemployment rate. This unrest was compounded by a high prevalence of HIV Tollefson et al.: Cleft Lip–Cleft Palate in Zimbabwe
infection and other infectious diseases, contributing to a life expectancy of below 40 years of age. During this period, the workforce lost a significant number of educated professionals and physicians due to emigration. The number of physicians in 2003 was noted to be approximately 1:10,000 (with even fewer subspecialist surgeons), in comparison to the United States with 19 physicians per 10,00013 (see Appendix A3 for additional geography, history, and healthcare indices of Zimbabwe). The prevalence of cleft lip–cleft palate in Zimbabwe has not been well documented. The incidence of cleft lip–cleft palate in the United States is often quoted to be approximately 1:600 to 1:750 live births15 (Appendix A4). Intermediate rates are seen in Caucasians (1:1,000 live births), whereas the lowest rates are seen within the African American community (0.3:1,000 live births).16–19 To relate this to African incidence rates, the incidence in Malawi has been described as 0.67:1,000 live births of cleft lip–cleft palate, but reports from Zimbabwe have been sparse.19 Therefore, to extrapolate from existing data, if the population of Zimbabwe is between 12 and 13.1 million in 2011, with a live birth rate of 27.7 live births per 1,000,20 and if we use the incident rate from the Malawi data of 0.67:1000, we would find approximately 217 live births with cleft lip–cleft palate per year. Even if the US rate of 1.67/1,000 live births were used, only 272 children with cleft lip–cleft palate would be born a year in Zimbabwe, a number that can reasonably be addressed. Of those live births, children with oral clefts, when compared to the overall population, have twice the odds of dying within the first year of life, which may be averted by implementing early interventions such as feeding and surgical services.21
Study Objectives The principle objective of this study was to evaluate the prevalence and unmet need for cleft lip–cleft palate reconstructive surgery by measuring the age of patient presentation. The incidence of cleft lip–cleft palate, acting as a proxy, can be used to estimate the unmet surgical burden of other congenital deformities. Our general hypothesis is that there remains a significant unmet need for surgical correction of congenital deformities. Evaluating the age of patient presentation may act as a proxy for overall disease burden, in that older patients, once treated, cause the age of presentation to decrease as the backlog of the surgical burden is managed. The older patients could be considered to have been treated by our team or related local surgeons, but other visiting teams were not present or very rare during the study period. A decrease in the overall burden may demonstrate a temporal trend in patient age at presentation. Our hypothesis was that the age of presentation to the biannual cleft team consultations will decrease in the overall patient population between 2006 and 2012. The secondary hypothesis was that the geospatial distribution of cleft lip–cleft palate may demonstrate clustering or expansion of catchment area to more rural areas (possibly due to expansion of communication about Laryngoscope 125: February 2015
the program, news coverage, and partnering). The research question asked is that compared to those children with isolated cleft palate, are those with cleft lip more likely to be presented to screening clinics due to the visibility of the cleft lip, as opposed to the less obvious palate defect? In the more rural areas, healthcare providers may not evaluate the palate as readily as an obvious cleft lip. Using geographic information systems (GIS) software, we sought to create epidemiological surveillance of where the children presenting to the cleft team screening clinics were travelling.
METHODS Study Design Two study designs are nested in this longitudinal crosssectional study. First, the epidemiologic aspect relies on estimating the prevalence of cleft lip–cleft palate from data on incidence and duration (age of patient). The second component involves estimating the geographic distribution of cleft lip–cleft palate prevalence in Zimbabwe for children presenting for evaluation to biannual surgery screening clinics in Harare, Zimbabwe between 2006 and 1012. Since 2006, the surgical team, Operation of Hope, has been making biannual surgical missions to evaluate, triage, and surgically correct cleft lip–cleft palate in over 600 patients. On the day of clinic presentation, the study sample was enrolled into the screening clinic using a standardized screening form (Fig. 1).
Database Creation: Study Population The screening form data were rigorously collected and entered into a secure, online database (Velos eResearch; Velos, Inc., Fremont, CA), which had been designed in conjunction with the University of California, Davis Biomedical Informatics, Clinical & Translational Sciences Center. Institutional review board approval was obtained through both the University of Zimbabwe School of Health Sciences and the University of California, Davis for this study. This method of capture is limited by including only those children who present for clinic evaluation.
Variables Study variables were based on the screening clinic forms. Our outcome measures included: continuous explanatory variables included age, birth weight, current weight, hemoglobin level, and travel distance in hours. Categorical variables were gender, diagnosis (isolated cleft lip, isolated cleft palate, cleft lip–cleft palate), diagnosis laterality, type of procedure, and revision surgery. Dichotomous variables included self-reported HIV status and tuberculosis status. The spatial variables were the home address, city, and province in Zimbabwe, which were converted to longitude and latitude.
Statistical Analysis Univariate analyses were conducted for all continuous variables to examine measures of central tendency using the Students t test and v2 test for categorical variables. Linear regression modeling was used to test for a time trend in patient age. The dependent variable was age of patient presentation, and the independent variable was the sequential year of recurring surgical trip. Analyses were conducted in R, version 2.13.0 (R Development Core Team, 2011; The R Foundation for
Tollefson et al.: Cleft Lip–Cleft Palate in Zimbabwe
S3
Fig. 1. Data entry form for each screening clinic for creation of the database (n 5 604).
Statistical Computing, Vienna, Austria). Statistical significance was considered as P .05.
Spatial Analysis: Application of Database Data to GIS Mapping Tool The Zimbabwe cleft lip–cleft palate data from 2006 to 2010 were incorporated into a spreadsheet format (Excel 2003; Microsoft, Redmond, WA). These data were converted for us in the GIS software (ArcInfo 10.0; ESRI, Redlands, CA). Geocoding was completed to obtain latitude and longitude coordinates for each patient’s home address based on the individual cities, towns, and closest village using Google Earth 7.1.1.1888. (Google Earth version 5.1.3533.1731, 2009; Google Inc., Mountain View, CA). Initially, 450 records were matched, but errors were discovered during the rematching. Abbreviated city names or provinces, misspellings, and street names were reviewed and corrected where possible. A total of 411 patients with cleft lip– cleft palate were captured. These data were used to construct maps for each year of presentation from 2006 to 2010 for all patients with a cleft (n 5 411). The geographic distribution of cases of cleft lip–cleft palate was created in sequential surgical team “mission/camp/brigade” based on month and year of the team arrival. For comparison, available population density maps were used as reference for where the largest population base is located in the country. Separate diagnosis-specific maps were created from longitude and latitude coordinates (cleft lip–cleft palate, isolated cleft lip, and isolated cleft palate) (Fig. 1). The distances and estimated travel times from home to the Harare screening clinics were calculated for the six most common locations.
Laryngoscope 125: February 2015
S4
RESULTS Patient characteristics are listed in Table I. Of the total 604 patients, 175 represented the combined cleft lip–cleft palate group. Isolated cleft palate was seen in 106 patients and isolated cleft lip in 251 patients. The median age for the cleft lip and cleft lip–cleft palate was between 1.5 and 2.0 years, whereas the isolated cleft palate patient’s median age at presentation was 4.2 years. The patient’s gender was evenly distributed in the cleft lip with and without cleft palate groups, and no difference was identified between the diagnosis groups (P 5.81). A slight male preponderance was noted in the isolated cleft palate group, with 50% male, 42% female, and 8% of unknown gender. Birth weight was significantly less for those with isolated cleft palate at 2.8 kg (standard deviation [SD] 0.5) compared to isolated cleft lip at 3.0 kg (SD 0.9) and cleft lip–cleft palate at 2.9 kg (SD 0.5). This suggestion of a lower birth weight for those with isolated cleft palate is still within the normal range and likely does not represent a significant clinical difference. The screening process only identified three patients with tuberculosis, and five were being treated for HIV. Preoperative hemoglobin levels (total mean 11.5 g/dL) were not different between diagnosis (P 5.398). The left-sided preponderance of cleft lip was seen to be present, with approximately 42% to 47% of the patients demonstrating a left-sided cleft lip and 23% to 34% as a right-sided cleft lip (Figs. 2 and 3). Tollefson et al.: Cleft Lip–Cleft Palate in Zimbabwe
TABLE I. Patient Characteristics by Diagnosis. Variable
Cleft Lip, n 5 251
Cleft Palate, n 5 106
Cleft Lip and Palate, n 5 175
4.9 (6.7) 1.5 (0.0–43)
5.5 (5.0) 4.2 (0.6–33)
4.9 (6.3) 2.0 (0.0–34)
3
—
1
Age, yr
No. missing Gender Male
6.3 (8.6) 3.0 (0.04–80) 5 .081
Female Unknown Birth weight, kg
122 (49%)
47 (44%)
82 (47%)
288 (48%)
118 (47%)
52 (49%)
90 (51%)
291 (48%)
11 (4%)
7 (7%)
3 (2%)
25 (4%) .025
Mean (SD)
3.1 (0.9)
2.8 (0.5)
2.9 (0.5)
3.0 (0.7)
3.0 (1.3–9.2) 117
2.9 (1.8–3.8) 47
3.0 (1.2–4.0) 39
3.0 (1.2–4.1) 247
16.6 (15.2) 10.0 (3.0–78)
17.2 (10.9) 13.8(4.6–62)
15.8 (13.9) 10.5 (2.9–70)
46
24
6
Current weight, kg Mean (SD) Median (range)
All Patients, N 5 604†
.250
Mean (SD) Median (range)
Median (range) No. missing
P Value*
.501
No. missing Diagnosis description Left
18.2 (15.9) 12.0(2.9–78) 95
