Need for HTA supported risk factor screening for hypertension and diabetes in Nepal: A systematic scoping review

Objective Health Technology Assessment (HTA) is a comprehensive and important tool for assessment and decision-making in public health and healthcare practice. It is recommended by the WHO and has been applied in practice in many countries, mostly the developed ones. HTA might be an important tool to achieve universal health coverage (UHC), especially beneficial to low-and-middle-income countries (LMIC). Even though the Package for Essential Non-communicable Diseases (PEN) has already been initiated, there is a clear policy gap in the HTA of any health device, service, or procedure, including the assessment of cardiovascular risk factors (CVRFs) in Nepal. Hence, we carried out the review to document the HTA supported evidence of hypertension and diabetes screening, as CVRFs in Nepal. Materials and methods We searched in PubMed, Cochrane, and Google Scholar, along with some gray literature published in the last 6 years (2016–2021) in a systematic way with a controlled vocabulary using a well-designed and pilot tested search strategy, screened them, and a total of 53 articles and reports that matched the screening criteria were included for the review. We then, extracted the data in a pre-designed MS-Excel format, first in one, and then, from it, in two, with more specific data. Results Of 53 included studies, we reported the prevalence and/or proportion of hypertension and diabetes with various denominators. Furthermore, HTA-related findings such as cost, validity, alternative tool or technology, awareness, and intervention effectiveness have been documented and discussed further, however, not summarized due to their sparingness. Conclusion Overall, the prevalence of DM (4.4–18.8%) and HTN (17.2–70.0%) was reported in most studies, with a few, covering other aspects of HTA of DM/HTN. A national policy for establishing an HTA agency and some immediately implementable actions are highly recommended.


Introduction
Health technology assessment (HTA) is a multidisciplinary approach incepted during the 1970s for assessing technologies such as drugs, devices, procedures, settings of care, services or programs, including the screening in health care with the collection of data from epidemiological, clinical efficacy, quality of life, service utilization, and cost (both health systems' and patients' out of pocket payments) studies.
HTA assists in evidence informed health policy making on the introduction and use of health technologies. Since any state or country has competing priorities, policy-makers often decide what is the best buy for the available evidence-base and the budget, which can further help as suggested by the WHO, in developing the health benefit package (HBP) for any state or country (1). At least two dimensions (if not three) of universal health coverage?the proportion of the cost and the services covered are directly linked to and implicated by HTA. Countries like the USA, Australia and Canada have been benefiting from HTA for decades, whereas Asian countries like South Korea, Japan, China, and India (recently) are gaining momentum (2). As per a survey carried out by WHO in 2015, Nepal has neither rendered a national HTA body nor is there clarity on how HTA is used in decision-making (3).
According to the global burden of disease study (GBOD) conducted in Nepal in 2017, non-communicable diseases were responsible for two-thirds of all deaths (66%), while high blood pressure and high fasting blood glucose levels were responsible for 14 and 10% of all deaths, respectively (4). Screening can avert disability adjusted life years (DALYs) of 2.33-3.1 in the case of hyperglycaemia (5) and is cost-effective in both hyperglycaemia and hypertension (5,6).
Diabetes is a metabolic disorder identified by screening using technologies like laboratory tests that include fasting plasma glucose, 2-h (2-h) post-load plasma glucose after a 75-g oral glucose tolerance test (OGTT); hemoglobin A1c (HbA1c); and random blood glucose in the presence of signs and symptoms. Diabetes is defined as fasting plasma glucose of 7.0 mmol/L (126 mg/dl), 2-h post-load plasma glucose of 11.1 mmol/L (200 mg/dl) (5), HbA1c ≥ 6.5% (48 mmol/ml); or a random blood glucose of ≥11.1 mmol/L (200 mg/ dl) in the presence of signs and symptoms. If elevated values are detected in asymptomatic people, repeat testing, preferably with the same test, is recommended as soon as practicable on a subsequent day to confirm the diagnosis (7).
Raised blood pressure can be identified at either a clinic, home-based, or ambulatory screening, and the diagnosis can be made with repeated measurements at 1-4 week intervals. In addition to this, blood tests, echocardiography, urine dipstick, albumin, and liver functions tests are other additional tests that are carried out in hypertensive suspects or diagnosed cases (8). As a part of assessing cardiovascular risk, mean systolic blood pressure should be measured over two separate occasions. For those who are already on anti-hypertensive medicine, the most recent recorded pretreatment value should be adopted as given by the National Institute for Health and Care Excellence/British Hypertension Society (NICE/BHS) guideline (9).
The population-based screening for diabetes and hypertension at a frequency ranging from every year to every 20 years was not cost-effective at the present level healthcare systems in India. However, it was estimated that screening at 3-to-5-year intervals could be cost-effective if the proportion of newly diagnosed and treated patients increased by more than 20%. Providing population-based screening of those two disease conditions through primary health centers (PHC) could be cost-effective; therefore they need to be strengthened at PHC level (10). A comparative study of BP measured in Shardaben General Hospital, India by digital and aneroid sphygmomanometers showed mean SBP as 108.92 ± 15.14 and 109.66 ± 16.81, and mean DBP as 76.20 ± 12.25 and 78.02 ± 14.35 mm of Hg, respectively, justifying that the bias for mean SBP and DBP was clinically non-significant and both instruments can be used interchangeably (11).
Diagnosing hypertension by sphygmomanometer is considered the gold standard as it has a high level of accuracy. A study from West Bengal showed that the aneroid device had better accuracy than the digital device as compared to the mercury sphygmomanometer (12). The mercury sphygmomanometer, which was widely used in the past to measure blood pressure in an office or clinic, has now been largely phased out in US hospitals, leading to the use of non-mercury, aneroid, or hybrid manometers in clinic and hospital settings (13,14). Ambulatory monitoring was the most cost-effective strategy for all ages of men and women after an Adhikari et al.
. /fcvm. . initial raised reading at the clinic or at home for blood pressure diagnosis, and so, it should be taken as a reference standard (15,16). Similarly, other instruments like 12-lead portable ECG devices for the screening of cardiovascular diseases in Ahemdabad found that screening at PHC by ECG saves 2. 9 lives per year at an incremental cost of 89.97 USD, yielding a cost-effectiveness ratio of 31.07 USD, so the facility to screen cardiac abnormality at PHC level is highly recommended for risk adults and symptomatic cases (17). It coincides with the US preventive services task force for screening for prediabetes and type 2 diabetes at ages 35-70 years in a primary setting, which has a moderate net benefit. It was suggested that screening every 3 years may be a reasonable approach for adults with normal blood glucose as the initial normal glucose test result is limited. The diagnosis of type 2 diabetes should be confirmed with repeated tests. Now, the US Preventive Services Task Force (USPTSF) changed their practice of 2015 recommendation and lowered the initial age of screening from age 40 to 35 among adults who were obese or overweight (18).
In Bhutan, an economic evaluation of the World Health Organization (WHO)'s Package of Essential Non-communicable diseases (PEN) found that the ambulatory but high-risk screening (where people who are overweight, obese, or >40 years for DM and/or HTN, visiting primary care facilities) represents good value for money compared to "no screening" (19). However, if performed on a regular basis and taking into account the specific population group and the existing non-disease to disease conversion rate, opportunistic high-risk approach screening may also yield high results (but may not confirm cases) (20).
The Indian diabetes risk score (IDRS) was used in a study from Tamil Nadu, India to determine the prevalence of type 2 diabetes high-risk cases. Use of the IDRS could reduce the cost of diabetes screening in India and so, being suggested to use in mass screening (21). Insulin pumps or glucose sensors appear costeffective, particularly in populations with higher HbA1c levels and rates of hypoglycaemia. However, the cost-effectiveness for combined insulin pumps and glucose sensors was less clear (22).
A study from Kuwait assessing type 2 diabetes and hypertension using machine-learning among 89,858 diabetics, 58,745 hypertensives, and 30,522 co-morbids shows that >85 and >90% accuracies were achieved for diabetes and hypertension, respectively using simple non-laboratory-based parameters. Talking more about the prediction, ethnicity is seen as a significant factor to the predictive models such as Asian-specific models and assessments perform even better (23). Population based screening for high-risk strategies can prevent cardiovascular disease. Also, it was suggested that where resources are limited, further advice be taken to take a total risk approach to identify several risk factors targeting to highrisk group to identify hypertension. Targeting high-risk groups for screening can help reduce costs because resources are not spent on the entire population. Early detection results in timely treatment and management of risk factors, which ultimately assists in reducing morbidity and mortality and reducing healthrelated costs (24).
A 540-person study in rural West Virginia was designed to identify diabetic risk using an HbA1c test and a diabetes risk assessment tool. Results showed that 61.8% of participants were at high risk for diabetes. It shows that community-based screenings are an effective way to assess diabetes risk (25).
Because cardiovascular risk factors were prevalent among Nepal's rural population, comprehensive intervention targeting all risk factors should be planned and implemented to reduce the burden of CVD in Nepal (26). Nepal has yet to work to implement health policies to tackle CVD or other NCDs. Recently, CVD management has been focused on treatment as there has been a rise in the availability of interventional cardiology and cardiothoracic surgery services (27). An assessment of health facilities for the implementation of a non-communicable disease package in Nepal among 92 health facilities in Kailali and Ilam districts revealed the gaps in the capacity of health institutions and the system in terms of training, supply, equipment, and diagnostics (28).
According to WHO, HTA should include the complete range of interventions or technology, and not be limited to just one. Technologies and methods considered in HTA include safety, clinical effectiveness, equity issues, ethical issues, feasibility, and acceptability of the health care system by providers and patients (29). Based on serum HbA1c levels, diabetes screening was offered at the Durham Veterans Affairs Medical Center for people aged 45-64 years old. Participants with an HbA1c ≥ 6.0% were invited for a follow-up measurement of blood pressure fasting plasma glucose, and health-related quality of life (HRQoL) were measured. Those with HbA1c ≥ 7.0% or fasting plasma glucose ≥7 mmol/dl are diagnosed as having unrecognized diabetes. There was no difference in the HRQoL of patients diagnosed with diabetes and those not diagnosed with diabetes. Screening for diabetes has minimal, if any, "labeling" effect with respect to quality of life (30).
Hypertensive disorders are the risk factors for cardiovascular diseases in Nepal, and there is a wide gap and socioeconomic disparity in hypertension management. In Nepal, the overall prevalence of high blood pressure was 19.6 percent. Furthermore, less than one-third of the population was treated, and <20% had their blood pressure under control. Wealth and education-based inequalities in awareness, treatment, and control measures of raised BP were significantly high in urban and rural areas (31). In this scenario, Nepal has developed PEN protocols and initiated the program in 2016 (32). These signify (but are not limited to) the systematic assessment of equity, cost, clinical efficacy, safety, ethically sound, feasible, and acceptability of health devices, procedures, and services before their population-wide applications.

Review framework
We used the following methodological framework to conduct this scoping review; (1) Identifying review questions, (2) Creating review objectives, (3) Establishing eligibility criteria, (4) Creating a search strategy and identifying search sources, (5) Screening records and data extraction, and (6) Evidence synthesis.

Search strategy
A comprehensive search was conducted using PubMed, Google Scholar, and the Cochrane databases, with publication date limits of the last 6 years (2016-2021). We also searched for reports, theses, and abstracts in gray literature and in citation searches.
A search strategy was finalized after developing a search strategy in consultation with the experts in the field. We searched with keywords such as screening, hypertension, diabetes, hyperglycaemia, Nepal, non-communicable, review, raised blood pressure, surveillance in the title, abstract, and keywords (ti; ab; kw); and with the medical subject heading (MeSH) descriptors such as heart disease risk factor, diagnostic techniques, blood pressure, diabetes mellitus. We searched on the basis of outcome indicators as illustrated in Table 1.

Study selection
After the strategy was finalized among all, a researcher, having trained and previous experience in systematic search and evidence synthesis, independently screened studies for eligibility and relevance. Studies were considered to be eligible if they had any of the outcome indicators as illustrated in .

Data extraction and data synthesis
Three different data extraction grids were developed by two review authors (CA and KS) and then sent to two authors (LMA and BP) for data extraction, from whom CA again obtained and compiled the data. Information regarding authors, country, study year, study type, sample size, geography, sex, age, ethnicity, and outcomes measures (Table 2), including key findings, and strengths and limitations (Table 3), was extracted from 53 individual studies (Figure 1), which were included in the review.

Results
After excluding duplicates, irrelevant studies, and further eligibility assessment, 53 articles were included in the scoping review.
Of the 53 included studies, 11 reported both DM and HTN prevalence. Participants aged 50 years and older from Bagmati Province and Terai had a higher prevalence of DM. Similarly, people aged 60-69 years from Karnali Province and urban hills had a higher prevalence of HTN. A higher DM prevalence was reported among the clinical participants. For HTN, it was higher among both healthy and clinical participants. Similarly, studies/screenings carried out in ambulatory settings had a higher prevalence for DM, but for HTN, it was reported to be higher in the camps. Males had a higher prevalence among participants with DM and HTN, and co-morbidity for       • This is the first study to report on trends and distribution of the CVD burden at a national level in Nepal.
• Prevalence of hypertension was 21% (JNC7) and 44% (2017 ACC/AHA) • Prevalence of hypertension awareness was 37.1 and 43.9% in male and female, respectively. • Prevalence of antihypertensive medication was 47.5 and 50.1% in male and females, respectively • Prevalence of control of hypertension among the hypertensive was 53.5 and 49.2% among male and female, respectively.
• Blood pressure was measured three times in a single day for the study whereas JNC7 guideline recommends longitudinal measurement • Possibly misclassification bias • Robust association between the outcome variables and caste could not be obtained due to missing data 9 Das Gupta et al.
• Overall prevalence of hypertension was 21.1% • Three blood pressure measurements were recorded; all were done in a single visit within a 5-min interval 10 Datta and Humagain (42) • Overall prevalence of prehypertensive and hypertensive women were 24.30% and 10.86 whose husband did not consume alcohol • 4.5% point gap in hypertension prevalence between wives of alcohol-consuming husbands and those of husbands not consuming alcohol • Likelihood of being hypertensive of Nepalese women was 12.8% • Husband's alcohol consumption, as a factor of wives' hypertension status.
• The most prevalent comorbidity of hypertension and diabetes was 5.7% followed by HTN and COPD (4.8%), and HTN and CKD (4%) • Secondary analysis of the data from the NCD survey 2018 • Chronic disease multimorbidity determinants and patterns study 12 Dhungana et al.  • Prevalence of hypertension in Nepal was 19.6%.
• Less than one-third (20.2%) of the hypertensive population received treatment and 10.4% among them had their blood pressure controlled. • The odds of being hypertensive was higher in men compared to women1.96 (1.59-2.44).
• Prevalence of hypertension was 7.1 (2.9-11.4) percentage points higher in affluent populations compared to the disadvantaged ones in Nepal.
• Provides detailed information on existing inequalities in prevalence and management of hypertension • The study cannot be generalized to population aged 50 years and above 22 Sainju et al. (53) • Pre-hypertension and hypertension were seen in 11.02 and 30.17% of the study population, respectively • Almost three-fifths of the obese participants were hypertensive • Sample is not nationally representative • Single episode of measurement of blood pressure (three readings) was taken, which may not be sufficient to diagnose hypertension in the population. • There could be an error due to observer variation in hearing the Koratkoff sound in crowed places 23 Saito  • Waist to height ratio and waist circumference were also included for picking up obesity with higher cardiovascular risk despite normal body mass index • Underlying causes and co-morbidities are not included 29 Silvanus et al. (60) • Prevalence of known diabetes (50/306) was an estimated 16.34% (95% CI: 12.62% to 20.90%) • 46.09% were classified as high risk, 44.53% as moderate risk and 9.38% as low risk for developing diabetes • The tabulated sensitivity (true positive rate) of the IDRS cut-off score ≥60 (high risk classification) was found to be 84.21% with a specificity (true negative rate) of 55.24% • The false positive rate and false negative rate was 44.76 and 15.79%, respectively.
• The positive predictive value was 20.0% and negative predictive value was 96.34% • Community-based study design to screen for undiagnosed diabetes, the step wise approach including the non-invasive tool and estimation of RCBG and the use of both FPG and the 2 h PG following a 75 g OGTT to identify diabetes and prediabetes • A community-based screening program can attract persons who have the health condition, those with a propensity to seek health care or who are more interested in their health which can introduce a selection bias 30 Silvanus et al. (61) • Prevalence of undiagnosed diabetes was 4.32% (95% CI 1.75-8.70%) and that of prediabetes was 7.14% (95% CI 3.89-12.58%) • The overall prevalence of persons with "raised blood glucose" was 11.73% (95% CI 5.64-21.28%) • All of the persons with prediabetes (n = 12) had IGT • Recognizing the use of glucometer and capillary sampling in low-and middle-income countries • Two glucometers were used during the screening camp, within glucometer variation was not studied 31 Tan et al. (62) • Most individuals with hypertension could link hypertension to its causes, symptoms and complications • Some individuals with hypertension occasionally stopped medication due to forgetfulness, negligence, laziness, and affordability issues • First qualitative study in Nepal involving a range of stakeholders to gather multidimensional insights into hypertension management • Qualitative design and small sample size limit the generalizability of the study findings 32 Tang et al. (63) • The range of 8.2-12.1% and 4.3-9.1% missed and overidentified hypertensive, respectively found when only 1st measurement was taken. • The range of 4.9-6.4% and 2.0-4.4% missed and overidentified hypertensive, respectively, found when only 2nd measurement was taken but for this, all the participants needed to go through screening. • The range of 3.8-8.1% and 2.0-3.9% missed and overidentified hypertensive, respectively, found when 2nd measurement was taken only for those having BP ≥ 130/80 during 1st screening. The range (%) of the participants needed to screen in the conditional screening (BP ≥ 130/80) for 2nd time is only 33.8-59.8% • Hence, resource cost is reduced by 40.2-66.2% when conditional sequential screening is carried out • Comparison of 1st, 2nd, and conditional 3rd screening was carried out so as to assess the difference in resources used • Survey data of USA is taken from 1999-2016 whereas findings from opportunistic screenings of May • Undiagnosed hypertension was disproportionately higher among lower socioeconomic status groups (Concentration Index, C = −0.18, p < 0.001).
• Nationally representative, cross-sectional data to determine the prevalence • Other behavioral and lifestyle factors potentially relevant to undiagnosed hypertension, for example, physical activity, dietary patterns and family history of hypertension, that were not explored in this study.

37
Kadaria and Aro (68) • 52% were moderately active • 28% were highly active • Facilitators and barriers physical activity were assessed • Self reporting and its recall bias on measure of physical activity 38 Karmacharya et al. (69) • Proportion of hypertension awareness among the hypertensives was 44.7% • Prevalence of taking antihypertensive treatment was only 76.1% (among the known hypertensives and 33.2% among the total) • Prevalence of control of hypertension was 35.3% among the known hypertensive and 11.7% of the total • Spectrum of awareness, treatment and control of hypertension • BP was taken in home setting and single day measurement cause false readings and affect the study • The targeted study was conducted at Dhulikhel which has teaching hopsital that could impact on level of awareness and control of HTN among the participants. 39 Khanal et al. (70) • Proportion of participants controlling Systolic BP increased to 58.3% from 3.3% compared to only to 40% among the intervention vs. control group • Percentage of the controlled Diastolic BP increased by 30% after the intervention compared to only 20% on usual care (control) • The study was study was conducted in one municipality and high number of female respondents thus limited generalization.  • Around 6% of participants had co-occurrence of two NCDs risk factors.
• Adjusted prevalence ratio (APR) was calculated from multiple poisson regression method • Secondary data analyzed for reproductive aged women 50 Neupane et al.
• Low, medium, and high levels of knowledge about hypertension were 43, 24, and 31%, respectively • No significant differences were observed in the knowledge and attitudes related to hypertension in relation to demographic characteristics of FCHV. • A majority of FCHV agreed that smoking (69.8%), alcohol (77.8%), low physical activity (42.4%), high salt intake (65.4%), high fat intake (78.7%), and genetics (53.9%) are major risk factors for hypertension.
• The study was conducted only among FCHV based in 1 municipality in Nepal 51 Neupane et al.
• HTN was 29.6%, M = 55.4%; F = 24.1 • Pre-HTN was 20.6% • The mean systolic blood pressure at 1 year was significantly lower in the intervention group than in the control group for all cohorts: the difference was −2·28 mm Hg (95% CI −3·77 to −0·79, p = 0·003) for participants who were normotensive, −3·08 mm Hg (−5·58 to −0·59, p = 0·015) for participants who were prehypertensive, and −4·90 mm Hg (−7·78 to −2·00, p = 0·001) for participants who were hypertensive • First cluster-randomized controlled trial to report systolic blood pressure among normotensive, prehypertensive, and hypertensive populations through an existing network of community health workers 52 Neupane et al.   both diseases ranged from 5.7 to 20.2%. Interestingly, pre-HTN and pre-DM were reported among about one-third of the participants, and importantly, for both DM and HTN, a minimum of one-fourth were unaware, and among those on medication, a maximum of only about half had their diseases under control (Table 4; Figure 2).

Validation and alternative use of the technology
The Indian Diabetes Risk Score (IDRS) is a simple tool that includes four statements to screen for diabetes and, with a cut-off score of 60, among the Nepalese population, revealed a sensitivity and a specificity of 84.2 and 55.2%, respectively, and hence could be a suitable alternative for diabetes and prediabetes screening (60). A hospital-based study revealed that measurement of serum Adenosine Deaminase (ADA), a biomarker, can be an alternative tool for glycaemic control and monitoring. It showed a significant correlation with HbA1c, fasting blood sugar (FBS) and post-prandial glucose (84). Similarly, a simple technological change in classification in the national guideline, changing the category of (130-140)/(80-90) mm of Hg as pre-HTN as defined by the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) to HTN as newly defined by the American Heart Association/American College of Cardiology (AHA/ACC), can reduce cardiovascular and related health problems by early detection and thereby save more lives and money (40). Another application of sequential screening, BP ≥130/80 mm of Hg, if assessed a second time (without assessing all for a second time), would give almost the same result but reduce the cost (time, human resources) by 40-60% during the camp and other opportunistic screenings (63). Moreover, the relationship of heavy metals with BP indicates that we need to address the problems of NCD prevention and control from a different perspective. On the other hand, HTN and DM are both webbed with multiple risk factors. For example, the triad of HDL-C, abdominal obesity and HTN was found among nearly 1 in 10 (8.2%) persons (76), and 6% women of reproductive age had co-occurrence of two NCD risk factors. This implies that these problems the rather be addressed with multiple and complex interventions, along with some feasible policy interventions (Figure 2).

Cost, DALYs, and equity
The prevalence of HTN was found to be more than seven (7.1) times higher among the wealthy than among the poor (31) but undiagnosed HTN was found to be disproportionate (1.6 times higher) among those with lower .
/fcvm. . socioeconomic status (C = −0.18, p 0.001) (67). According to a qualitative finding, 52% of rural and 63% of urban people considered the medicine used to treat HTN and DM to be expensive (50). A review of costing studies showed that in DM, USD 445.9 (±27,535), 16.2 and 5.1 are needed per annum per patient, per prescription, and per visit per patient, respectively (48), thus, implicating medication non-adherence ( Figure 2).

Awareness, medication adherence, and intervention e cacy
Nine out of every 10 people have moderate to high chances of developing DM (60), which indicates the importance of early interventions. HTN medication was missed due to forgetfulness, negligence, and unaffordability (62), which could be attributed to a lack of or poor HTN knowledge among community members (50) and the female community health volunteers (FCHVs) (81). However, when these cadres were trained and prepared as change agents, the intervention could reduce a mean blood pressure by 2.3-4.9 mm Hg and increase the proportion of BP control and knowledge of the community members (70, 82).

Discussion
Overall, the prevalence of high blood pressure was found to be as low as 5.7% (43) and as high as 70% (46) nationally, and up to 55% (39) in Karnali province and in urban areas. Similarly, DM was found to be between 5.7 (43) and 15% (46) nationwide, though it was as low as 6.9% (39) and as high as 32.5% (52) in Karnali and Bagmati provinces, respectively. Similarly, there is also a wide variation in DM (45-94%) and HTN (9-95%) awareness (49,56,58,72), disease/disorder controlled by medicine (56, 58, 71, 77) in the population. This variation should be considered when planning and implementing the relevant programs for coverage and efficacy (Figure 2). A combined study of systematic review and expert consultation carried out in eight European countries found that the incorporation of additional social value judgments beyond clinical benefit assessment and economic evaluation could help further explain heterogeneity in coverage .
/fcvm. . recommendations and decision-making (85). Similarly, it would be wise to compare the cost in lieu of the yield and the diagnoses that the screening strategies give. In an ambulatory high-risk approach carried out in the Bhutan PEN evaluation study (19), a 10% of eligibility and around 23% of diagnosis for HTN and a 10% of eligibility and 26% of diagnosis for DM may be contrasted with a 13% of eligibility and 17% of diagnosis for HTN and a 19% of eligibility and 22% of diagnosis in an opportunistic high-risk screening approach carried out in Karnataka, India (20). In these two studies, number of diagnoses was higher in ambulatory highrisk approach but yield was higher in opportunistic highrisk approach. Producing HTA capacity in a country like Nepal may run up against different problems at various stages, such as policy sensitization, development and implementation, expertise development, and overall management. A similar program carried out by the Health Intervention and Technology Assessment Program (HITAP) of Thailand in India, Colombia, Myanmar, the Philippines, and Vietnam revealed experiences suggesting that it is not only technical capacity, such as analytical techniques for conducting economic evaluation, but also management, coordination, and communication capacity should be strengthened (86). Inequality in HTN was observed to have up to 1.6 times higher prevalence among the poorest and underprivileged (31, 67). A combined study of documentary review and interviews carried out by Ciani et al. in Italy during the early stage of HTA development, argued that not only the central agency should be in place, but also a fully coordinated and harmonized multi-level structure of HTA was imperative (87). Then, the Italian National Healthcare System was one of the most decentralized systems since the devolution reform approved in 2001, and Nepal is more or less, at present, in a similar situation with devolution of power to the local levels with a 3-tier governing system.
Regarding the universal health coverage (UHC) and social health security program (SHSP) that are underway, we are adopting and scaling-up different benefit packages by assessing only a few to many, but not all components. The early results of Thai universal coverage in 2009-10 showed that HTA is helpful for informing coverage decisions for health benefit packages because it enhances the legitimacy of policy decisions by increasing the transparency, inclusiveness, and accountability of the process (88). The National Institute for Health Research HTA programme also suggests that synthesized evidence, rather than depending upon clinical trials, should be taken into account, for decision analysis and that the findings of a trial be linked to systematic reviews and metaanalyses (89).

Limitations
The included studies' principal limitations were costing comparisons of technologies in light of comparative validity and precision. Furthermore, there was a lack of ethical, ICER, quality of life, and willingness and capacity to pay for various technologies.

Conclusion
Overall, 53 studies, including some secondary analysis and the reviews, mainly reported the prevalence of DM of 4.4-18.8% and HTN of 17.2-70%. In addition to establishing an HTA national agency, some immediate actions and a systematic HTA of both diseases, covering a representative sample, is highly recommended before warranting.