Few shot chain-of-thought driven reasoning to prompt LLMs for open ended medical question answering