OBJECTIVES: This study examines trade-offs between the high dimensionality of covariate matching and high computational efficiency from propensity score applications by comparing efficiency among six distinct hybrid algorithms used with healthcare data. METHODS: Six matching algorithms were examined. Each combined covariate matching with a different propensity scoring function: continuous factor, weighting factor, caliper, parenting factor, nesting factor or partner. The algorithms were compared in terms of 1:1 matching rate, computing time, bias balancing and standardized difference. The influence of sample size variation on stability and efficiency was considered. Paired T-test, Pearson Chi-Square and Standardized Difference were adopted for assessment. RESULTS: The superiority of some hybrid algorithms over pure covariate matching was observed. In terms of matching rate, the partner function reported the highest rate (99.7%), followed by its function as a caliper (88.4%), while the parenting function produced the lowest rate (59.5%). All others performed at a similar level. Computing time varied, the most efficient using the propensity score as a parenting factor (00:25:10). The longest reported times were seen when used as a weighting factor (00:37:56) or caliper function (00:37:52). Differences are more profound in large samples. In bias balancing tests, all algorithms were balanced on categorical covariates except when the propensity score was used as a partner or a caliper where each displayed the lowest capability of producing p-values above 0.05. Significant reduction in standardized difference below 10% was indicative of higher efficiency of the hybrid algorithms. Categorical covariates produced values near zero despite the lower performance for the partner approach. With increasing sample size, all investigations performed as expected. CONCLUSION: Overall, these hybrid applications exhibited greater efficiency in simultaneously overcoming high dimensionality on covariate matching and reducing variation in propensity score matching. Depending on data characteristics and research profiles, each application has specific merits in certain circumstances.