Hypertension is a chronic disease with high prevalence and a major public health problem due to its severe consequences and associated costs. It is estimated that 24% of Colombians suffer from this disease (1). Hypertension is linked to an increased cardiovascular risk, which manifests as complications such as acute myocardial infarction, stroke, heart failure, angina, and death (2). According to clinical practice guidelines (2-4), some patients may require two or more active substances to achieve adequate control of blood pressure; however, there is an inversely proportional relationship between the number of medications administered over a period of time and adherence to treatment.

Various studies have documented that adherence to antihypertensive treatment decreases as the number of daily doses and the count of active substances administered separately increases (5-7). In this context, fixed-dose combinations of antihypertensives, i.e., two or more active substances in the same pharmaceutical form, have been proposed as a strategy to improve adherence (8), which in turn could result in a higher probability of achieving therapeutic goals and a reduction in the risk of cardiovascular complications associated with hypertension. However, the institutional cost of fixed-dose antihypertensive combinations is significantly higher than that of free combinations with the same active substances (9). This raises questions about their efficiency as a public health intervention. Therefore, it is necessary to determine whether the use of a fixed-dose combination represents a cost-effective alternative compared to the free combination (FC) of the same active substances.

To date, no previous studies in Colombia or Latin America have evaluated the cost-effectiveness of using fixed-dose combinations compared to free combinations of the same active substances. Economic analyses published with similar research questions have been conducted in other jurisdictions with significant differences in population, drug markets, and health systems, which makes it difficult to transfer their results to the Colombian context (10-13). Based on the above and considering that the combination of losartan with hydrochlorothiazide is one of the most frequently prescribed antihypertensive combinations, both in its fixed and free forms in Colombia, an economic evaluation is justified to determine the cost-effectiveness of using fixed-dose combinations of losartan plus hydrochlorothiazide versus free combinations of the same active substances for achieving therapeutic goals in patients over 18 years of age with hypertension.

An economic evaluation of cost-effectiveness was conducted from the perspective of the Colombian health system. The objective was to compare the use of fixed-dose combinations versus free combinations of losartan with hydrochlorothiazide, using as the effectiveness variable the proportion of patients who achieve blood pressure readings lower than 140/90 mmHg. Since this measure of effectiveness is typically observable within the first year of treatment, a one-year time horizon was defined, a period during which death or significant changes in quality of life are less likely to occur, and for this reason, they were not included in the study. Likewise, it was not necessary to apply a discount rate.

Blood pressure control was chosen as the primary effectiveness outcome, considering that it is the direct clinical mediator of the impact of adherence in the short term, a period during which changes in quality of life are usually not sensitive to generic utility instruments. The decision tree structure (62 probabilistic nodes and 80 terminal nodes) responds to the need to exhaustively capture the clinical trajectories of patients, conditioned by three levels of decision-making: therapeutic adherence level, achievement of therapeutic goals, and the presence of specific adverse events (hypokalemia, hyperkalemia, and hyponatremia).

Although the effectiveness outcome is binary, the inclusion of multiple terminal nodes allows for the assignment of differentiated costs based on the type of metabolic complication and its clinical management, thereby ensuring that the incremental cost-effectiveness ratio (ICER) accurately reflects the economic burden from the perspective of the third-party payer.

The effectiveness data and probabilities were obtained after a thorough review of the medical literature, and were subsequently validated by a clinical expert. In line with the recommendations of the Institute of Health Technology Evaluation of Colombia (14), the medication prices were extracted from the Medication Price Information System as of 2023. All values were converted to the minimum concentration unit (MCU) and expressed in Colombian pesos. Additionally, the costs of medical care for managing adverse events were obtained from the Individual Health Service Provision Records database.

The uncertainty of the model parameters and the robustness of the results to potential variations in the values used were assessed through both deterministic and probabilistic sensitivity analyses. The deterministic analysis was performed univariately on the model parameters, as this approach identified which variables had the greatest impact on the ICER. Meanwhile, the probabilistic sensitivity analysis considered the simultaneous uncertainty of all parameters (multivariate) through Monte Carlo simulations (1000 iterations), assigning beta distributions to the probabilities and gamma distributions to the costs, according to the nature of each variable. This approach revealed the dispersion of results in effectiveness versus cost graphs and constructed acceptability curves that allowed for evaluating the effect of different willingness-to-pay thresholds on the simulations. This facilitated decision-making based on the results, especially when there is no defined threshold in Colombia for the effectiveness variable defined in this study.

Assumptions of the model include: 1) patients have the same probability of receiving each intervention, 2) patients experience only one adverse event at a time, 3) adverse events are not conditioned by each other, 4) patients do not experience adverse events other than those described, 5) patients do not receive medical management different from the specified one, 6) adverse events are attributed solely to the intervention, 7) the intervention remains unchanged during the evaluated period, and 8) no patient dies from an adverse event.

A total of 518 studies were reviewed by title and abstract, of which 463 were excluded. It was found that the fixed-dose combination (FDC) of losartan with hydrochlorothiazide is associated with higher therapeutic adherence compared to its free combination (35.4% vs. 28.2%). Furthermore, greater blood pressure control was observed in patients who adhered to treatment, compared to those who did not (67.7% vs. 55.3%); however, there was a slightly higher incidence of adverse events (2.20% vs. 1.28%). The most frequent adverse event was hyperkalemia, occurring in 4.95% of patients, followed by renal function decline in 4.45%, and hypokalemia in 1.08% of patients.

Additionally, a higher average price was confirmed for the FDC of losartan with hydrochlorothiazide compared to the free combination (COP 1.47 vs. COP 1.44 per MCU); however, greater variations in the maximum prices were found (COP 2.08 vs. COP 70.53 per MCU), which can be explained by the brand-name medications, which show a greater price dispersion. The average cost of adverse events during the time horizon was COP 758,595 for hyperkalemia, COP 589,363 for hyponatremia, COP 314,605 for hypokalemia, COP 275,952 for angioedema, and COP 126,381 for renal function decline.

The base case results showed that FDCs had a higher annual cost per patient than free combinations, with average values of COP 26,991 compared to COP 17,084, respectively. However, the costs incurred led to greater effectiveness, as the proportion of patients who achieved blood pressure control was 0.589 with FDC compared to 0.580 with free combinations, generating an incremental cost-effectiveness ratio (ICER) of COP 1,164,588 per additional patient achieving hypertension control.

The deterministic sensitivity analysis, shown in Figure 1, demonstrated that the cost of the FDC, the probability of adherence to the free combination, and the probability of control in adherent patients were the variables that generated the most uncertainty, with ICER variability between their minimum and maximum values of COP 146,321,070, COP 20,135,325, and COP 2,510,154, respectively. In contrast, the effect of adverse events on the ICER was comparatively insignificant, as it generated variability of COP 118,088 in adherent patients and COP 87,978 in non-adherent patients. On the other hand, the Monte Carlo simulation showed inconsistencies in the results due to the high standard deviation in the prices of brand-name medications, which produced extreme parameters for generating random numbers within their respective probability distributions; for this reason, the study proceeded by stratifying based on generic medications.

Figure 1

FDC: fixed-dose combinations; FC: free combinations; ICER: incremental cost-effectiveness ratio.

The results of this study show that generic fixed-dose combinations (FDC) of losartan and hydrochlorothiazide are cost-effective compared to the free combinations (FC) of the same active substances for achieving blood pressure control in patients with hypertension. This difference is mainly explained by higher therapeutic adherence observed with the FDC, which translates into a higher probability of achieving blood pressure control within the one-year time horizon. Although the incidence of adverse events was slightly higher in the adherent group, this increase appears to be more related to the longer exposure time to the medication than to a real difference in the safety of the alternatives compared. Nevertheless, the clinical benefit of achieving controlled blood pressure far outweighs the impact of these adverse events, especially due to their low frequency of occurrence in the model. A relevant finding from the sensitivity analysis is the low influence of adverse events on the model's stability. Although managing complications such as hyperkalemia or angioedema involves significant institutional costs, their low probability of occurrence in the treated population dilutes their impact on the ICER calculation. Figure 2 confirms that the uncertainty in the costs of these events does not alter the direction of the result.

From an economic perspective, the FDCs showed higher annual costs, but also superior effectiveness, which results in an ICER of COP 1,584,163 per additional patient achieving blood pressure control. This figure is below the hypothetical willingness-to-pay threshold of COP 2,040,816, obtained through the acceptability curve, at which the probability that the FDCs are cost-effective consistently exceeds the probability for FCs. This suggests that, in the Colombian context, this intervention may be considered cost-effective.

These findings are consistent with international studies that have demonstrated that FDCs of antihypertensives improve adherence, persistence, and blood pressure control (15-17). Furthermore, they align with the results of other economic evaluations where FDCs were found to be cost-effective compared to FCs of antihypertensives within the willingness-to-pay thresholds defined in the countries where the studies were conducted (12,13). However, these evaluations used metrics for their populations; they did not stratify by generic and brand medications due to the minimal availability of generic drugs in their local context; they employed quality-adjusted life years (QALYs) as an effectiveness measure, and implemented longer time horizons that allowed for the evaluation of outcomes such as cardiovascular complications and death. Nevertheless, this study offers valuable insight for the local context, as, despite the existence of an economic evaluation for antihypertensives in Colombia from 2023 (18), it does not compare FDCs versus FCs of equivalent active substances. Moreover, it does not include stratification by generic and brand medications.

The evidence suggests that in a chronic disease like hypertension, where pharmacological management is long-term and adherence is key to therapeutic success, implementing strategies that promote treatment compliance should be a priority. Based on these results, it is suggested that decision-makers consider prioritizing the use of generic FDCs, as long as their price remains within reasonable margins, as a mechanism to improve clinical outcomes in the hypertensive population and, in the long term, reduce the economic burden of cardiovascular diseases.

This study has some important limitations, including the exclusive use of secondary sources to populate the model, the exclusion of individual clinical variables such as age, comorbidities, or sex, and the methodological decision to use a one-year time horizon, which limits the inclusion of major cardiovascular events or long-term outcomes. Additionally, a single constant treatment was assumed for each patient during the year, and the high price variability of brand-name medications affected the validity of the probabilistic analysis, leading to its exclusion. Another relevant methodological consideration is the use of an intermediate outcome, rather than final utility measures like QALYs. This decision is based on the asymptomatic nature of hypertension and the one-year time horizon. Evidence has shown that intensive blood pressure control does not lead to significant changes in the quality of life reported by patients or their general well-being in similar follow-up periods (19).

Therefore, the cost-effectiveness analysis presented here is more sensitive and relevant for institutional decision-making than a cost-utility analysis, which would require long-term Markov models to capture major cardiovascular events.

Finally, among the limitations, the exclusion of brand-name combinations introduces a selection bias that limits the generalization of the results to the entire market. This decision was necessary due to the behavior of the brand data, whose high variability generated simulations with extreme values that distorted the model's results. However, it is likely that the impact of this bias is minimal, since the institutional market in Colombia, seeking efficiency, is predominantly focused on the use of generic medications.

Despite the limitations, multiple opportunities for future research were identified, including the use of primary data, the implementation of more complex models such as Markov, subgroup clinical analyses, the inclusion of long-term clinical outcomes, and the evaluation of regulatory strategies on institutional prices. Additionally, the methodology can be replicated for other antihypertensive combinations or chronic diseases such as diabetes.

Generic fixed-dose combinations (FDC) of losartan with hydrochlorothiazide have proven to be cost-effective for achieving therapeutic goals in hypertension compared to the free combinations (FC) of the same active substances. The acceptability curve showed that, starting from the hypothetical threshold of COP 2,040,816, the probability that the FDC is cost-effective is higher than that for the FC. Therefore, the decision to adopt this alternative is at the discretion of the decision-maker, who should evaluate, in light of the Colombian context, the possibility of assuming the additional costs of its adoption.

The findings and recommendations of this study can serve as input to guide decisions on health policies and the management of antihypertensive treatments in Colombia, with the aim of preventing cardiovascular complications, improving the quality of life of patients, and reducing the costs associated with the disease. Although the use of generic medications and price regulation have been part of the country's public policy for decades, the findings of this study highlight the need to periodically review the specific behavior of FDC prices compared to FCs and ensure that coverage decisions are based on current and transparent economic evaluations for a more efficient use of health system resources.