I'm a postdoctoral level biomedical researcher who has used many other biostatistics books before. This book is currently used for entry-level biostatistics at my School of Public Health. The bonus is that this book is geared towards epidemiology (eg, there is a chapter on rates and standardization), and I liked the CDROM that comes with the book and contains datasets.
Unfortunately, the authors have chosen to derive many key concepts from probability theory. This is fine if you are comfortable with probability concepts. However, most students with background in social sciences are not: thus I found that this approach often adds an unnecessary level of complexity.
Most of the examples are long-winded, dragging on from page to page - flipping back and forth is extremely frustrating, and the layout makes it difficult to grasp what the point of each example is. Also, the long-winded style makes this book a poor desk reference if you just quickly need to check a formula or definition. Sadly I have to say that I very rarely use this $100 book - there are plenty of more accessible and succinct entry-level texts that cost a fraction of this volume. And yes, Google is better learning tool than this book.