3

Data Availability and Gaps
in Four Countries

The country discussions summarized in Chapter 2 emphasized the importance of accurate data and noted limitations in current data collection. An additional session on the second day of the workshop, with representatives from Bangladesh, Kenya, Grenada, and Rwanda, was designed to further explore the sorts of data that are most needed to support a decision-making process, the currently available data sources in their countries, and the specific challenges their countries face with respect to data. Reliable and accurate information is a critical resource for decision makers, session moderator Stephen Jan of the George Institute for Global Health observed. A useful toolkit would need to provide data to guide decision makers in “institutionalizing a process of rational resource allocation,” Jan commented, meaning not just a process of exploring cost effectiveness but a process for using evidence to identify the optimal ways to meet the objectives for a particular country’s health system.

Jan noted that certain types of data are particularly likely to be useful: health care expenditures; cost offsets (savings that might accrue as a result of particular expenditures); social costs and benefits, such as gains or losses in productivity resulting from health problems or health interventions; the cost effectiveness of interventions and resources; costs that are relevant to health outcomes but that are borne by other sectors, such as education, housing, or labor; and resource capacity, including infrastructure and human resources. It will also be important, he added, to identify needs and the ways they may vary across population groups, regions, classes, and disease group, as well as to understand community preferences for the allocation of resources. There are many potential sources for such data, Jan said, in-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 29
3 Data Availability and Gaps in Four Countries T he country discussions summarized in Chapter 2 emphasized the importance of accurate data and noted limitations in current data collection. An additional session on the second day of the workshop, with representatives from Bangladesh, Kenya, Grenada, and Rwanda, was designed to further explore the sorts of data that are most needed to sup- port a decision-making process, the currently available data sources in their countries, and the specific challenges their countries face with respect to data. Reliable and accurate information is a critical resource for decision makers, session moderator Stephen Jan of the George Institute for Global Health observed. A useful toolkit would need to provide data to guide deci- sion makers in “institutionalizing a process of rational resource allocation,” Jan commented, meaning not just a process of exploring cost effectiveness but a process for using evidence to identify the optimal ways to meet the objectives for a particular country’s health system. Jan noted that certain types of data are particularly likely to be useful: health care expenditures; cost offsets (savings that might accrue as a result of particular expenditures); social costs and benefits, such as gains or losses in productivity resulting from health problems or health interventions; the cost effectiveness of interventions and resources; costs that are relevant to health outcomes but that are borne by other sectors, such as education, housing, or labor; and resource capacity, including infrastructure and hu- man resources. It will also be important, he added, to identify needs and the ways they may vary across population groups, regions, classes, and disease group, as well as to understand community preferences for the allocation of resources. There are many potential sources for such data, Jan said, in- 29

OCR for page 29
30 COUNTRY-LEVEL DECISION MAKING cluding research studies, health ministry data, hospital data, insurance data, and international resources, but such data may not all be of equal quality. “We need to identify gaps and data needs,” he said, “but also to identify the tradeoffs between quality and pragmatism.” BANGLADESH Bangladesh is fortunate in having a number of sources of data on chronic diseases, said Tracy Pérez Koehlmoos, of the ICDDR,B, although she added that they are “perhaps not collected with the rigor we would like to see.” Sources include reporting data from public hospitals; specialty hospital reports; a survey of risk factors and burdens in six cities (Angeles et al., 2008); standardized data collected using the World Health Organi- zation’s (WHO’s) Stepwise Approach to Surveillance (WHO-STEPS)1 and published research literature. These data are valuable but do not provide a complete picture, Koehl- moos said. There are no national registries for cancer, for example, al- though cancer specialty hospitals and medical schools do report some data, which are included in a biennial health report. The specialty hospitals, she added, are “quite strong” but have mostly “worked in isolation rather than thinking about [chronic diseases] as a common issue that needs to be addressed by the whole health system.” The Urban Health Survey data, although they do not cover the whole country, are quite useful because Bangladesh’s population is heavily concentrated in urban areas. Koelhmoos expressed the hope that the next round of the Demographic and Health Survey will provide more data on chronic diseases and risk factors. Also, there has also been a recent “explosion” of primary and secondary studies related to noncommunicable diseases in Bangladesh, she added. Bangladesh also has health and demographic surveillance sites run by the ICDDR,B.2 For 40 years, ICDDR,B has followed the lives of more than a quarter of a million people, collecting data about their health, em- ployment, marriages and divorces, and other indicators, Koehlmoos said, which has helped to document the rapid growth in Bangladesh’s mortality rates for noncommunicable diseases. In 1986, these diseases accounted for approximately 10 percent of all deaths in Bangladesh, but by 2006, they accounted for approximately 70 percent of deaths (Karar et al., 2009). Bangladesh has many programs of various types to combat chronic 1 For more information about WHO STEPS, see http://www.who.int/chp/steps/en/ (accessed October 2011). Koehlmoos and presenters also cited the value of the INDEPTH network of surveillance sites; see http://www.indepth-network.org/ (accessed October 2011). A participant noted that the WHO STEPS program in Bangladesh was particularly easy to set up, in part because of the presence of the INDEPTH network. 2 See http://www.icddrb.org/ (accessed October 2011).

OCR for page 29
31 DATA AVAILABILITY AND GAPS IN FOUR COUNTRIES diseases, including programs operated by specialty hospitals and national associations as well as camps for groups who have these diseases. However, Koehlmoos said, these programs have not been rigorously evaluated, so there is little to say about their effectiveness. It is similarly difficult to be precise about the economic burden of chronic diseases, she added, because few studies have been done in this area. Koehlmoos also noted that although the country has national strategic plans for both cancer control and the control of noncommunicable diseases, these plans have not been fully implemented. “What we are seeing is long delays between thinking about how to address an issue and actually getting to the point where action is taken to deliver services and create awareness,” she said. Another problem is a lack of consistency between the private and the public sector; furthermore, private-sector providers have no requirement to report their data. Even within the public sector there is a lack of continu- ity. For example, urban areas fall under the Ministry of Local Governance rather than the Ministry of Health and Family Welfare, and there is not yet unified reporting across these units of government. However, the Ministry of Health and Family Welfare has recently created a directorate devoted to noncommunicable diseases, which Koehlmoos believes will improve coor- dination of resources, although funding is still an issue. Of the directorate’s total 2008-2009 budget of $10.2 million, nearly $9 million was channeled to a program dealing with arsenic poisoning (a problem that currently af- fects approximately 20 million people in Bangladesh) and to strengthen the Institute of Public Health (for example, by purchasing equipment for public hospitals). Moreover, even where data are collected there are gaps. For example, Koehlmoos noted that the Urban Health Survey collects data on tobacco use among males but not among females. The reason is that smoking is viewed as extremely inappropriate for women in Bangladesh, so few of them are thought to smoke; nonetheless, the precise numbers cannot be known if the data are not collected. Several research efforts now under way are expected to yield data on chronic diseases soon, including a national survey of risk factors for noncommunicable diseases; the WHO-STEPS research; and a study of salt intake, public beliefs about salt, and policy related to salt usage that was funded by United Health Group. Several studies funded by the U.S. National Heart, Lung, and Blood Institute are also under way, focusing on health-seeking behavior and health systems, prevalence and determinants of chronic obstructive pulmonary diseases, prevalence of hypertension and risk factors, and chronic disease and poverty. Research on healthy aging in Bangladesh and the cost of prevention of noncommunicable diseases is also forthcoming (El Saharty et al., 2011; Mirelman et al., 2012). Koehlmoos closed by offering her sense of the most urgent needs for

OCR for page 29
32 COUNTRY-LEVEL DECISION MAKING targeting chronic diseases in Bangladesh. One such need is to perform a rapid assessment of barriers to moving from the “pre-action phase to ac- tion.” It will also be very important to evaluate existing programs, at least to assess how many people are being reached, and to assess the availability of drugs and services for treating these conditions. Koehlmoos also sug- gested that the Ministry of Local Health and the Ministry of Health and Family Welfare devise a unified reporting system and that noncommuni- cable disease risk factors be included in future Health and Demography Surveys. There are models that have been shown to work well in Bangladesh that can be emulated as the country moves forward in a campaign against chronic diseases, Koehlmoos said. One such model is offered by the Ex- panded Programme of Immunization, in which the ministries of defense, education, and health collaborate to make sure that young people have necessary vaccinations. Public–private partnerships have also been success- ful, such as in the way in which nongovernmental agencies worked with the government in efforts to reduce tuberculosis. Koehlmoos suggested that a first step toward dealing with chronic diseases would be to convene a national platform that would engage all potential partners in conduct- ing an inventory of existing data, performing a comparative analysis of risk factors, and carrying out an economic burden analysis. It will also be important to carry out a qualitative analysis of implementation issues, she said, and together such information could support effective planning and programs. KENYA As Gerald Yonga suggested in his presentation summarized in Chapter 2, there are a number of problems with data collection in Kenya. In his later presentation on data availability and gaps, Yonga reviewed a range of materials developed by the Kenyan government, professional societies, researchers, and international organizations in order to develop a picture of publications and data related to chronic diseases. Looking across this material, he found that it supplies a good amount of epidemiological in- formation, such as data about disease prevalence and incidence, but very little information or analysis of costs and few economic burden analyses. Even less information is available about interventions for chronic disease in Kenya, and there is virtually nothing known about the costs of interven- tions. “Regional data predominates,” Yonga said. Primary national data are almost nonexistent; policy makers must instead rely on “estimates ac- cumulated from various sources.” As discussed in Chapter 2, the cost and prevalence data that are avail- able indicate that chronic diseases are an increasing problem in Kenya. For

OCR for page 29
33 DATA AVAILABILITY AND GAPS IN FOUR COUNTRIES example, although there is no national cancer registry, hospital registries and studies collectively provide a reasonable picture of the significant can- cer burden and the rates of particular cancers. “There is room for strength- ening” the available data on risk factors and screening for these diseases, Yonga noted, in order to obtain a more representative national picture, but the existing information does provide a useful basis for action. Both the direct costs of treating and managing patients and the loss of productive years of work from chronic diseases are significant, Yonga noted, but there are few economic studies of either. Much of the available information comes from studies that included Kenya as one of a number of countries, such as the WHO Choice study and the Intersalt study of the cost savings that can occur when a population reduces its salt intake (Murray et al., 2003; Rose et al., 1988). Furthermore, there are few studies of the cost effectiveness of interventions, either those targeting individuals or those targeting populations. Such studies could provide useful information for determining health care policies in Kenya, but their lack of country-specific information is often very apparent. For example, one study using regional data recommended certain guidelines for risk modeling of cardiovascular disease that relied heavily on the results of laboratory tests, such as for blood sugar and cholesterol. However, it was impossible for Kenya to use this type of modeling and to use blood sugar and cholesterol tests as mass screening tools because the country does not have the laboratory resources and testing policies needed to collect this data (Lemogoum et al., 2003). A study that had focused only on Kenya might have produced more helpful guidelines for screening programs and risk detection that were less labora- tory-based and that instead took advantage of the resources that Kenyan health centers can access. To summarize, Yonga identified some of the principal gaps in Kenya’s chronic disease data: • National data on the prevalence and incidence of noncommuni- cable diseases and risk factors • National surveillance of diseases and risk factors • Data on the economic burden of these diseases and their costs at the community and household level • Evidence on feasible and cost-effective risk modeling, screening methods, and intervention programs for cardiovascular and other chronic diseases • Collection of health information from private-sector hospitals and other health activities • Access to unpublished data (e.g., from nongovernmental projects or university theses) • Connection between evidence needs and research activities

OCR for page 29
34 COUNTRY-LEVEL DECISION MAKING Yonga suggested several steps that Kenya could take to fill in some of these gaps. Primarily, it could implement the WHO-STEPS survey and simultaneously work on enhancing existing data and health information systems, including e-health systems, in a way that would allow the country to capture data on noncommunicable diseases. For example, Yonga sug- gested that the Kenya Demographic Health Survey, as a regular instrument with an existing logistical structure, has the potential to be used to collect national representative data, but it does not currently include a lot of non- communicable disease indicators. In addition, noncommunicable disease screening and intervention programs should be integrated into existing HIV and maternal and child health systems, and these programs could then be equipped with mechanisms that would make it possible to do costing and economic analysis. GRENADA While total deaths in Grenada have declined in the last 5 years, the number and percentage of deaths caused by chronic diseases has increased during that time, said Emma Herry-Thompson, Chief Medical Officer of Grenada’s Ministry of Health. Unfortunately, she added, the data available on interventions are “minimal.” As Francis Martin noted in the presenta- tion summarized in Chapter 2, surveillance data in Grenada consist pri- marily of weekly tallies from community health nurses. Hospitals supply medical records (of births, diagnoses of discharged patients, and deaths), and emergency rooms supply records of accidents and violence. The gov- ernment’s infectious disease control department also collects weekly data on those diseases. This is a paper-based system, and, as Herry-Thompson explained, the individuals responsible for collecting and reporting the data are not always committed to the importance of the task and thus data are frequently lost. A number of other sources of data are being developed, Herry- Thompson said, including the WHO-STEPS survey, WHO’s Global Youth Tobacco Survey,3 the Global School-Based Health Survey of the U.S. Cen- ters for Disease Control and Prevention (CDC), the Grenada Heart Project, and research by the Retina Resources Foundation of New York. However, she expressed concern that inefficiency and gaps in data collection may limit the value of some of these results. She also noted that project sustain- ability is often a problem. The Retina Resources Foundation did a great deal of surveillance and screening and even treated some patients with diabetic retinopathy, but the program was not sustained. Herry-Thompson 3 See http://www.who.int/tobacco/surveillance/gyts/en/ and http://www.cdc.gov/gshs/ (ac - cessed October 2011).

OCR for page 29
35 DATA AVAILABILITY AND GAPS IN FOUR COUNTRIES expressed the greatest optimism about the Grenada Heart Project.4 In this case, researchers from the World Heart Federation conducted a survey to profile risk factors in Grenada using the WHO STEPwise approach for the surveillance of risk factors. This work demonstrated a pressing need for interventions, some of which are in now in progress. Finally, Herry-Thompson noted that much of the data in Grenada is regionally driven, submitted as a requirement to regional surveillance organizations like the Pan American Health Organization (PAHO) and then used to generate regional reports that are fed back to the Ministry of Health. Herry-Thompson is hopeful that improvement in country-driven generation and use of data is under way. Herry-Thompson described the primary source of health care financ- ing in Grenada as general taxation, though support from other nations and international organizations such as PAHO, WHO, the United Nations Children’s Fund, and the Caribbean Development Bank are also important sources. However, the external funds are generally specifically targeted in ways that may not perfectly match the needs the ministry perceives. One result of the limited funding has been to prompt Grenada’s health officials to shift their focus to prevention, she said. Grenada’s biggest need, in Herry-Thompson’s view, is for a monitor- ing and evaluation system that could help ensure that limited resources are spent in optimal ways. The country also needs technical support to as- sess costs and cost-effectiveness, she said. Grenada has insufficient trained personnel for data collection and analysis, and it also needs both software and hardware for the collection and analysis of data. Public–private part- nerships need to be strengthened to facilitate data collection and sharing, she added, and the country needs a chronic disease registry. On the posi- tive side, Herry-Thompson concluded, there is a growing awareness of the importance of data among health officials and workers. Grenada has some challenges with its health infrastructure, but there is a good supply of well- distributed facilities. RWANDA It is not essential to have complete population-based data about non- communicable diseases in order to act, noted Gene Bukhman of the Har- vard Medical School and Partners in Health. In many countries with very low levels of current health spending, he said, “there is probably sufficient data to know we should be doing more for problems we are already aware 4 Seehttp://www.world-heart-federation.org/press/news-old/demonstration-projects/grenada- heart-project/print.html (accessed October 2011).

OCR for page 29
36 COUNTRY-LEVEL DECISION MAKING of.” Bukhman described the primary sources of health data for Rwanda as well as some concerns related to that data. The Demographic and Health Survey is one key source, he said. It provides various data relevant to chronic diseases, such as the fact that 93 percent of rural Rwandans live in housing with a dirt floor, versus 52 percent of urban Rwandans, and that 19 percent of urban woman have a BMI over 25, versus 10 percent of rural women (Institute National de la Statistique du Rwanda and ORC Macro, 2005). However, Bukhman said, it is important to place data on the prevalence of risk factors and diseases in context. For example, any response to data on tobacco use should take into account that Rwandans who smoke typically have only a few cigarettes per day, and that studies show that the number of cigarettes smoked per day has a major effect on disease outcomes. Similarly, although obesity is a significant health concern, undernutrition, malnutrition, and starvation are more serious conditions, and should lead to greater concern in a country such as Rwanda. This prompted a discussion among participants at the workshop about whether obesity and malnutrition are competing health issues. Gerald Yonga pointed out that undernourished and obese popula- tions exist side by side in many places. “With increasing urbanization, you see very underprivileged conditions where the problem is not complete lack of food but lack of choice.” One participant suggested that “unless we are able to show that the two need to be discussed at the same time, we have the risk of the entire noncommunicable disease agenda just being dropped lower on the priority list.” Another source of information is the estimates of the global burden of diseases that have been produced by WHO5 (Mathers et al., 2008), which indicate that noncommunicable diseases make up 25 percent of the disease burden in Rwanda. Such data do provide a general sense of the magnitude of the relative burden, but Bukhman described a few concerns pertaining to these data. For instance, the estimates were not based on data for particular conditions from Rwandan populations because such data are not available. Rather, the data are extrapolated from other sources, and thus may not be representative of what is happening in Rwanda. There may, for example, have been a “major overestimation of the burden of ischemic heart disease in Africa and for rural African sites,” Bukhman said. One participant com- mented that many countries face a similar problem, with burden of disease statistics being presented that sound serious but that are not based on any data from the country itself. Population surveys are also useful sources of information, Bukhman said, but the low prevalence in Rwanda of some of the chronic diseases has limited the collection of such data. The only nationally representative 5 See http://www.who.int/topics/global_burden_of_disease/en/ (accessed October 2011).

OCR for page 29
37 DATA AVAILABILITY AND GAPS IN FOUR COUNTRIES population surveys conducted in Rwanda have been for epilepsy, severe musculoskeletal impairment, and severe visual impairment (Atijosan et al., 2008; Mathenge et al., 2007; Simms, 2008). Thus, those studying chronic disease in Rwanda look to data from similar populations and settings for an indication of what is probable in Rwanda. Bukhman described an ex- ample of how data from rural communities in Gambia could be used to draw conclusions about the situation in Rwanda. Data from Gambia show that, while rates of mild hypertension are significant, only 1 to 2 percent of the population has severe hypertension (blood pressure over 180). The data from Gambia also indicated that most cases of hypertension are not accompanied by such other risk factors as obesity, diabetes, or high cho- lesterol (van der Sande et al., 1997). If one assumes that the risk factor profile would be similar in Rwanda, Bukhman suggested, this information could influence decisions about priorities in services and the development of care protocols. One source of information is data from facilities, which indicate the relative importance of particular diseases as a cause of hospitalization. “It’s very important to have the correct denominator” when considering such data, Bukhman said. For example, if one included obstetric admissions in the data, “it will make the burden of [chronic] diseases in terms of admis- sions and deaths seem very small.” Another complication is if the coding used to record diseases is not standardized, which increases error. Record- keeping such as health facility registers was interrupted during the 1990s by the Rwandan genocide, Bukhman said, but gaps are now being filled in retrospectively. An increasing number of districts now have electronic medical recordkeeping systems. Electronic medical records and monitoring of indicators will be key “in better documenting the risk factors and dis- ease patterns that affect Rwandans, which will make it possible to improve intervention effectiveness,” Bukhman said. Some data concerning the cost of interventions are available, he said, but most of the data concern actual procurement costs—the actual outlays for drugs, for example. The resulting figures are rough, he added, but they do provide a general picture, as Table 3-1 shows. These figures reflect spending on transport costs, community health worker and nurse salaries, hospitalization, and the like, which were used to calculate a per capita increase in spending, as well as a total. Bukhman noted that interventions for conditions that are very expensive to treat may be cost-effective if the treatments are effective, in part because very few cases of some of them are diagnosed. Another information source is the joint annual work plan. All of the parties that contribute to health care in Rwanda report exactly what they contribute and for what purposes, Bukhman said. It may not be completely accurate, but it provides a useful general picture. Looking forward, Bukhman identified several promising opportunities

OCR for page 29
38 TABLE 3-1 Rough Cost Data for Selected Chronic Diseases in Rwanda Average Annual Cost Case-Finding Population Condition Rate Prevalence Cases Found Total Per Patient Per Capita Cardiomyopathy 30% 0.2% 6,000 $2,009,506 $334 $0.20 Cardiac surgical follow-up 3% 0.1% 300 $3,709 $412 $0.012 Screening and follow-up for 100% 0.1% 9,000 $46,873 $5 $0.005 HIV nephropathy Diabetes 50% 0.44% 22,000 $3,806,655 $173 $0.38 Hypertension (160/90 100% 4% 400,000 $7,632,542 $9 $0.76 threshold) Chronic respiratory disease 15% 2% 24,900 $1,453,695 $57 $0.14 Chronic care integration $3,655,917 $1.50 subtotal Cardiac surgery (initial surgery) 3% 0.1% 300 $1,164,000 $3,880 $0.12 TOTAL $819,917 $1.62 SOURCE: Kidder et al. (2011).

OCR for page 29
39 DATA AVAILABILITY AND GAPS IN FOUR COUNTRIES for improving the data situation in Rwanda. One is a forthcoming study of rheumatic fever prevalence, which will influence the decision on whether or not to treat sore throats at community clinics. Another is the possibility of taking greater advantage of the mobile clinical experiences many young doctors have by using them to collect clinical population data. The devel- opment of a national cancer registry would also be very valuable. Finally, Bukhman mentioned other opportunities, such as the WHO-STEPS surveys, the use of sentinel surveillance sites, and the International Network for the Demographic Evaluation of Populations and Their Health in Develop- ing Countries (INDEPTH), although there are challenges to implementing these methods. Overall, he said, it is clear that there is a need for collecting population-level clinical data. SUMMATION From the presentations across all four countries, there emerged a clear need for strategies to collect and analyze data to assist an evidence-driven process for allocation of resources for chronic disease control. As Stephen Jan mentioned, a process for decision making regarding resource allocation can be aided by data that exposes the disease situation on multiple levels, including the basic disease epidemiology; whether there is a need or prefer- ence for more intervention; the social and economic disease burden; the available resources that can be used for disease control and the capacity to absorb new investment; the short- and long-term costs, benefits, and effec- tiveness; and the potential for sharing some of the cost burden with other sectors outside of health. Collecting an abundance of data to address each of these areas is a challenge; however, some data sources are available and do provide evidence that can serve as a basis for action. According to the reports from Bangladesh, Kenya, Grenada, and Rwanda, the majority of available data addresses disease epidemiology and comes from regional surveys and studies run by multilateral organiza- tions such as WHO, public and private hospital data, and various academic research studies. Some countries also conduct local surveys, some have surveillance sites that provide data, and some have national health manage- ment information systems that aggregate data from many facilities. There are, however, many gaps in the epidemiological data—several countries have no national registries for certain chronic diseases, others lack continu- ous and systematic reporting methods, some are dealing with weak commu- nication between the government and private hospitals or companies that could provide health data, and others lack data on disease risk factors. A common theme across countries is the lack of country-specific data—many countries are using what they can from regional studies, but the recommen-

OCR for page 29
40 COUNTRY-LEVEL DECISION MAKING dations from these studies are not always aligned with a country’s available resources and community preferences. The representatives of Bangladesh, Kenya, Grenada, and Rwanda de- scribed less availability of information regarding the social and economic burden of disease or the costs and benefits of using various resources. The speakers mentioned that there are few country-specific studies on the ef- fectiveness of programs targeting chronic diseases, the cost effectiveness of interventions, or the economic burden of disease. This is in large part due to the fact that there are few people in the health sector with the techni- cal skills to do these assessments. Several speakers mentioned that there is limited availability of services and interventions in their countries, and it is difficult to evaluate the effectiveness of something that is not present in the first place. Overall, the speakers suggested that the data that are available provide a general, even if not completely comprehensive, picture of the chronic dis- ease situation in the countries that they represent. It is clear that action for chronic disease control is needed, and the current data sources already pro- vide some guidance for decision making on what measures should be taken. As many of the participants mentioned, data gaps should not be an excuse for inaction; however, better data will ideally lead to smarter spending and more effective programs that effectively address the chronic disease burden. Chapter 6 provides a further summary of the considerations raised in this session along with the presentations and discussions throughout the workshop.