warns that no p-value is computed, but p-value is still computed

url="https://raw.githubusercontent.com/tompollard/data/master/primary-biliary-cirrhosis/pbc.csv"
data_pbc=pd.read_csv(url)

# 6 categories of age based on decade
data_pbc['age_group'] = data_pbc['age'].map(lambda x: int(x/10))
columns = ['age_group', 'age', 'sex', 'albumin', 'ast']
categorical = ['age_group']
table = TableOne(data_pbc, columns=columns, categorical=categorical, groupby='sex', pval=True)
table

gives:

tableone.py:463: UserWarning: No p-value was computed for age_group due to the low number of observations.
  warnings.warn('No p-value was computed for {} due to the low number of observations.'.format(v))

but the p-value still appears in the output table

variable	level	isnull	f	m	pval	ptest
n			374	44
age_group	2	0.0	3 (0.8)		0.022	Chi-squared (warning: expected count < 5)
age_group	3		63 (16.84)	3 (6.82)
age_group	4		116 (31.02)	10 (22.73)
age_group	5		121 (32.35)	14 (31.82)
age_group	6		62 (16.58)	13 (29.55)
age_group	7		9 (2.41)	4 (9.09)
age		0.0	50.16 (10.24)	55.71 (10.98)	0.001	One-way ANOVA
albumin		0.0	3.49 (0.42)	3.53 (0.46)	0.536	One-way ANOVA
ast		106.0	122.63 (57.92)	121.99 (47.01)	0.949	One-way ANOVA

tompollard / tableone

warns that no p-value is computed, but p-value is still computed #39