OBJECTIVES:  Dose-response meta-analysis (DRMA) is widely employed to establishing potential dose-response relationship between exposure and disease outcome. However, no method is readily available for exploring relation between discrete exposure and a binary or continuous outcome.   METHODS: We proposed a piecewise linear (PL) DRMA model which provide a solution to this issue. We used parity and sleep data to illustrate how to apply PL model in DRMA for assessing relation between discrete or continuous exposure with outcome. We also empirically compared PL model with nonlinear spline model. RESULTS: PL model fitted well in our two examples. For parity and risk of rheumatoid arthritis (discrete exposure): among women with 3 or less birth, the RR was 0.88 (95%CI: 0.77, 1.00) for every 1-birth increment; otherwise (3 or more births), the RR was 1.10 (95%CI: 0.99, 1.23) for every 1-birth increment. For sleep duration data: RR of all-cause mortality was 1.31 (95%CI: 1.25, 1.39) for every 1-hour reduction of sleep duration among people who slept less than 7 hours; and was 1.15 (95%CI: 1.07, 1.24) for every 1-hour increase among people who slept more than 7 hours. For continuous exposure, the results of PL model were less precise and flexible compared to higher order function. CONCLUSIONS: Piecewise linear function is a simple and valid method for DRMA and can be used for discrete exposure. It also represents an alternative to non-linear model DRMA.