“…These models allowing for two separate types of zeros are known as zero‐inflated count models (Mullahy, , Lambert, ), the most prominent ones being the zero‐inflated Poisson (ZIP) and zero‐inflated negative binomial (ZINB) models. In health economics, they have been used, among others, for the number of physician visits (Yen et al ., ; Sarma and Simpson, ; Sari, ; Pizer and Prentice, ), the number of pharmacy visits (Chang and Trivedi, ), the number of prescriptions (Street et al ., ), the number of occupational injuries (Campolieti, ), and the number of cigarettes smoked (Sheu et al ., ; Bauer et al ., ).…”