Table 3 Characteristics of the most recent studies narratively discussing the impacts of P4P for diabetes management by patients’ and/or providers’ sex/gender
Author, year | Study location | Population | Comparisons | Outcomes measured | Study analysis method | Sex-disaggregated reporting |
|---|---|---|---|---|---|---|
LeBlanc et al., 2016 [36] | Canada (New Brunswick) | Adult patients with diabetic glycosylated hemoglobin profile followed by a fee-for-service physician | Patients with/without physician uptake of incentives | − Number of hemoglobin A1c tests − Mean hemoglobin A1c levels | Linear and logistic regression mixed models of linked administrative and laboratory blood test records | − Sex of the patient − Sex of the physician |
Lippi Bruni et al., 2009 [37] | Italy (Emilia-Romagna) | Adult patients with type 2 diabetes based on diagnostic profile | Patients with/without physician uptake of incentives, by the presence/absence of a regional P4P scheme | − Hyperglycemic hospital emergency admissions | Multilevel modeling of linked administrative health and hospital records | − Sex of the patient − Sex of the physician |
Iezzi et al., 2014 [6] | Italy (Emilia-Romagna) | Adult patients with type 2 diabetes based on drug utilization and specialized care referral profiles | Patients with/without physician uptake of incentives, by the presence/absence of a regional P4P scheme | − Hospitalization for long-term diabetes complications: renal, eye, neurological, and circulatory disorders − Hospitalization for short-term diabetes complications: diabetic ketoacidosis, hyperosmolarity, and coma | Poisson regression models with fixed and random effects specifications of linked longitudinal health administrative records | − Sex of the physician |
Yuan et al., 2014 [38] | Taiwan | Adult patients with type 2 diabetes having participated in a clinical evaluation program under P4P | Patients’ length of participation in a diabetes education program | − Diabetes self-management practices − Changes from baseline in hemoglobin A1c levels | Multilevel linear regression modeling of longitudinal program records | − Sex of the patient |
Hsiesh et al., 2017 [39] | Taiwan | Patients with type 2 diabetes based on diagnostic profile with comorbid cancer | Patients with/without physician enrolment in P4P | − All-cause mortality − Diabetes-related mortality − Cancer mortality | Multiple regression analysis with propensity score matching of case and control cohorts of linked administrative health records, deaths registry, and cancer registry | − Sex of the patient |
Pan et al., 2017 [40] | Taiwan | Patients with newly diagnosed type 2 diabetes based on diagnostic profile | Patients with/without physician enrolment in P4P | − Physician Continuity of Care Index (COCI) − All-cause mortality | Multiple regression analysis with propensity score matching of case and control cohorts of linked administrative health records | − Sex of the patient |
Crawley et al., 2009 [41] | United Kingdom (England) | Adults reporting physician-diagnosed diabetes, heart disease, or hypertension | Patients’ occupational group | − Hemoglobin A1c, blood pressure, and cholesterol levels − Use of medications | Multiple regression analysis of annual household survey data including interviews and direct physical measures | − Sex of the patient |
Millet et al., 2009 [42] | United Kingdom | Adult patients with type 1 or type 2 diabetes according to medical records | Patients with/without selected comorbid conditions | − Hemoglobin A1c, blood pressure, and cholesterol levels | Multilevel modeling of longitudinal primary care records from a representative sample of general practices | − Sex of the patient |