Do not insert duplicate data

I'm a bit confused, what would be the best practice to not insert duplicate data in a table, I have the following code.

cnn = pyodbc.connect ( 'Driver={SQL Server};Server=**;Database=**;uid=**;pwd=**',autocommit=True )


sqlcmd = ("Select GEOM.STX as lng,GEOM.STY as lat,a.Codigo from Cie a \
                       inner join Inmueble b on a.idInmueble = b.idInmueble \
                       inner join DiviPolitica c on c.idDiviPolitica = b.idDiviPolitica \
                       inner join Dire d on d.idDivisionPolitica = c.idDivisionPolitica where d.geom is not null")

    read = fiona.open(r"\geo-pre\data\ViasActivos.shp")        

    base = cnn.cursor()
    base.execute(sqlcmd)

                for feature in read:
                    coord = feature['geometry']['coordinates'][0]
                    name = feature['properties']['Name']
                    linea = feature['properties']['LINEA']
                    ramal = feature['properties']['RAMAL']
                    cabecera = feature['properties']['CABECERAS']

                    for r in base:
                        longVias = coord[0]
                        latVias = coord[1]

                        cordenadas = '\''+str(longVias)+','+str(latVias)+'\''

                        center_point = [{'lat': latCIE, 'lng': lngCIE}]
                        test_point = [{'lat': latVias, 'lng': longVias}]

                        lat1 = center_point[0]['lat']
                        lon1 = center_point[0]['lng']
                        lat2 = test_point[0]['lat']
                        lon2 = test_point[0]['lng']

                        radius = 5.00 # radio en kilometros

                        a = haversine(lon1, lat1, lon2, lat2)

                        insertVia = ("INSERT INTO [dbo].[atributosCapas] \
                                                                  (idCapaOrigen,[idTipo] \
                                                                  ,[idOrigen] \
                                                                  ,[GeomTipo] \
                                                                  ,[Geom] \
                                                                  ,[Fecha] \
                                                                  ,[Nombre] \
                                                                  ,[Direccion] \
                                                                  ,[Localidad] \
                                                                  ,[Provincia],descripcion,cadena)\
                                        VALUES (null,3,2,'Punto',null,GETDATE(),'%s',null,null,null,null,'Ramal:%s,Linea:%s,cabecera:%s','%s')" % \
                                 (name,ramal,linea,cabecera,cordenadas))

                        base.execute(insertVia)   

The first for goes through a shapefile, the second for goes through an MSSQL query that I later insert into a table, now the only field I have to ask if that data already exists in the table [dbo]. [attributesCapas] is by the name (name), which would be the best practice for not inserting duplicates

Edit: I add to the question that I would like to avoid techniques like NOT IN or EXIST